Fatigue is an important ageing mechanism for long-term operation (LTO) of nuclear power plants. The effect of the reactor coolant environment on fatigue is one of the factors to take into account. Application of environmental fatigue codes, generally, leads to large margins for actual fatigue failure of components. In combination with longer operation times, these margins can make it challenging to meet the criteria defined in the standards. The purpose of the work in this paper is, therefore, to gain insight into the conservatism in typical fatigue analyses using design fatigue curves which is done by analyzing the fatigue process up to the actual component failure using crack growth analyses.
Prediction of fatigue failure in cylindrical specimen is investigated through numerical simulations of fracture mechanics. A finite element analysis model is implemented for a crack in a cylindrical specimen typically used for fatigue life testing and generating design fatigue curves. The specimen is uniaxially displacement-loaded into the plastic regime, and the reaction force is evaluated as a function of crack depth and shape. Cyclic loading leads to formation of a crack with the depth such that a failure point in the S-N curve corresponds to the 25% load drop. Stress intensity factors are calculated, and number of cycles to failure are determined based on Paris’ law and a two-stage crack growth relationship. Simulation results are compared to experimental fatigue life data and show good agreement. The outcome of the investigation can be extended to fatigue life of geometrically complex thermo-mechanical components, such as nozzles, under transient thermal loading occurring during operation of a nuclear power plant, in order to assess conservatism of fatigue failure criteria based on experimentally obtained S-N design curves.