Multiscale damage criterion for fatigue life prediction in metallic materials

In this paper a multiscale damage criterion for fatigue life prediction is introduced. A damage evolution rule based on strain energy density is modified to incorporate crystal plasticity at the microscale. A damage tensor that indicates the local damage state is derived using optimization theory. Local damage information is passed from local to grain level in the form of a damage vector via averaging techniques. Finally, the damage evolution rule for a meso representative volume element (RVE) is calculated through the Kreisselmeier- Steinhauser (KS) function, which can produce an envelope function for multiobjective optimization. The weighted averaging method is also used to obtain the corresponding damage evolution direction for the meso RVE. A critical damage value is derived in this paper to complete the damage criterion for fatigue life prediction. Simulation results and comparison with experimental data are presented.