At present, the fatigue life of nuclear reactor components is estimated based on empirical approaches, such as stress/strain versus life (S∼N) curves and Coffin-Manson type empirical relations. In most cases, the S∼N curves are generated from uniaxial fatigue test data, which may not truly represent the multi-axial stress state at the component level. Also, the S∼N curves are based on the final life of the specimen, which may not accurately represent the mechanistic time-dependent evolution of material behavior. These discrepancies lead to large uncertainties in fatigue life estimations. We propose a modeling approach based on evolutionary cyclic plasticity that can be used for developing finite element models of nuclear reactor components subjected to multi-axial stress states. These models can be used for more accurately predicting the stress-strain evolution over time in reactor components and, in turn, fatigue life. The model parameters were estimated for 316 stainless steel material, which are widely used in U.S. nuclear reactors. The model parameters were estimated for different test conditions to understand their evolution over time and their sensitivity to particular test conditions, such as the pressurized water reactor coolant condition.

This content is only available via PDF.
You do not currently have access to this content.