Fatigue crack propagation in friction stir welded joints significantly affects aircraft structural integrity. This study investigates the influence of welding speed, rotational speed, specimen thickness, loading frequency, and stress ratio on the fatigue crack growth rate. Four classical machine learning models with different structures—Deep Back-Propagation Network, Random Forest, Support Vector Regression, and K-Nearest Neighbors—were employed to predict fatigue crack growth behavior. The results show that all models achieve strong predictive performance. For FSWed joints, Deep BP and KNN exhibit comparable performance (R² > 0.98) on the training data, indicating similar learning capabilities with sufficient data coverage. Notably, KNN achieves the fastest training time (<0.3 s), while all models require less than 5 s of computation time. These results demonstrate that machine learning-based models provide an efficient and reliable alternative for rapid fatigue crack growth evaluation, supporting damage-tolerant design and structural integrity assessment in aircraft engineering.

The integral welded panel represents a highly promising aircraft structural component, owing to its lightweight design and reduced connector requirements. However, the complexity of its welded structure results in the formation of cross-welded joints. This study systematically investigated the mechanical properties of the cross-welded joints through tensile tests across different welded regions, which were complemented by fracture morphology examination via scanning electron microscopy (SEM). The residual stress distribution was characterized using X-ray diffraction, while electron backscatter diffraction (EBSD) analysis was used to elucidate the relationship between residual stress and microstructure. Key findings revealed that the cross-welded zone exhibited lower yield strength and ductility than the single-welded zone, and the advancing heat-affected zone demonstrated superior tensile properties relative to the retreating side. Residual stress analysis showed that the cross-welded joint lacked the “double peak” profile characteristic and displayed lower maximum residual stress than the single-welded joint. EBSD analysis indicated significant grain elongation in the cross-welded zone due to mechanical forces during the welding process, resulting in higher dislocation density and deformation, corresponding with elevated residual stress levels.