A microstructure sensitive modeling approach for fatigue life prediction considering the residual stress effect from heat treatment

Abstract

A multiscale numerical method to study the high cycle fatigue (HCF) and very high cycle fatigue (VHCF) properties of bearing steels is proposed in this study. The method is based on the microstructur sensitive modeling approach resulting from the integrated computational materials ensfginerrring concept, and further consider the effect of residual stress generated from the prior heat treatment processes. The microstructure features, including the grain size and shape distribution and inclusion size and shape description, are represented by the representative volume element (RVE) models. The matrix mechanical response to the cyclic loading is described by the crystal plasticity (CP) model. The CP material parameter set is calibrated inversely based on the strain-controlled low cycle fatigue tests. The results show that the residual stresses, especially those around the inclusion, have a great effect on the fatigue properties, which provides the key factor to give the correct prediction of the fatigue crack initiation site.