Study and Methodology Development for Cyclic Loading Application to Fatigue Analyses

This study concerns collection, review and development of cyclic loading methods applicable to probabilistic fatigue induced crack growth analyses. These approaches are applicable to nuclear power plant (NPP) piping components.
The main characteristics of cyclic loading are the frequency of occurrence and the amplitude. Both of these can remain constant or vary, the latter option describing much more realistically actual cyclic loading conditions than the former. There are several cases of cyclic loading with locally highly varying frequency of occurrence and amplitude. This is typical for such NPP piping Tees, where turbulent mixing of two water flows having different temperatures occurs. Such high-cycle loading conditions with substantially varying temperatures call for probabilistic more realistic method for application of loading, instead of applying such conservative application methods where some severe constant frequency of occurrence and amplitude are assumed. The use of such computationally simpler and more straightforward methods leads to unrealistically short fatigue crack growth lifetimes. The scope of this study mainly concerns thermal high-cycle loading. This study covers and reviews the most relevant application methods for cyclic thermal loading given in codes and rules of a number of countries using nuclear energy, as well as most promising methods found from fitness-for-service procedures and scientific literature. The developed new application method for cyclic loading is compared to a corresponding existing one in a computational example concerning fatigue induced crack growth in a representative NPP piping Tee. To obtain a more distinctive model response, moderately exaggerated loading was used. The computational fatigue crack growth analysis results show that the new method is considerably less over conservative than the SIN-Method.