Abstract
This research tackles the fundamental issue of computational fatigue studies by developing an effective approach that combines the crystal plasticity finite element method (CPFEM) with the Tanaka–Mura–Wu (TMW) model for crack nucleation and the Tomkins model for fatigue crack propagation, to provide Class-A predictions of the total coupon-fatigue life (crack initiation and growth lives) for a nickel-based superalloy, Haynes 282. To gain a statistical significance accounting for the microstructure inhomogeneity, 11 3D Representative Volume Elements (RVEs) are created utilizing Dream.3d to represent the polycrystalline material with different grain structures and orientations in equivalence to the experimental microstructure data. The CPFEM model is calibrated to the material's hysteresis behavior, and then, the microstructural plastic strain from the RVE is taken to calculate the fatigue life. The prediction is found in good agreement with the fatigue test data, validating the effectiveness of the proposed approach in predicting the fatigue life and scatter due to microstructural variability for Haynes 282 alloy. In addition, the effects of local grain attributes including grain orientation and adjacent grain arrangement on fatigue crack nucleation are analyzed quantitatively. It is suggested that grain orientation influences plastic deformation by inducing the active slip systems, and the slip transfer across grain boundaries also contributes to fatigue crack nucleation.