A Workflow to Accelerate Microstructure‐Sensitive Fatigue Life Predictions

Microstructure‐sensitive crystal plasticity simulations can be employed to augment experimental data and facilitate decisions for the insertion of new materials or manufacturing processes into the design cycle of components. While there has been tremendous progress in virtual microstructure instantiation, formulation, uncertainty quantification, and validation of microstructure‐sensitive fatigue modeling over the years, these processes can be time‐consuming in their development and implementation, especially using finite element solvers for crystal plasticity (CP‐FEM). This work explores a workflow to accelerate microstructure‐sensitive fatigue life predictions, leveraging the computationally efficient elastoviscoplastic fast Fourier transform crystal plasticity solver, instead of more time‐intensive CP‐FEM techniques. The prediction of fatigue life for additively manufactured IN718 is used as a case study to test the proposed framework. Different microstructure model configurations are analyzed to quantify how different voxel resolutions, representations of key microstructure features (e.g., annealing twins), and applied loading cycles affect accuracy. The predicted lives are then benchmarked against the predictions of a traditional CP‐FEM framework, and the accuracy is related to the reduction in implementation and computational time, analyzing the potential trade‐offs between precision and time efficiency to enable faster decision‐making regarding fatigue performance.