A Fatigue Short Crack Path Prediction Criterion Based on Improved Energy Release Rate

ABSTRACT Fatigue crack path prediction is critical in structural failure analysis. However, short crack growth, influenced by microstructures, cannot be accurately predicted by conventional fracture mechanics. This study proposes a new criterion based on an improved energy release rate, incorporating a stored energy density threshold and a crack driving state function into the maximum energy release rate criterion. Using crystal plasticity constitutive relations, a numerical model for GH4169 nickel alloy is developed and validated via fatigue tests and SEM. The simulation and experimental results show that the new criterion significantly improves path prediction accuracy compared with traditional methods, especially in the prediction of the crack deflection angle at the grain boundaries, and the average deviation of the proposed criterion from the experimentally observed values is less than 5°. This enhances prediction reliability and supports structural fatigue life assessment.