Summary on effect of initial flaw and load assumptions on risk estimate changes

This report presents a summary of a three years long study on the failure probabilities of nuclear power plant piping components. The focus of the study has been on effect of initial flaw and load assumptions on nuclear power plant (NPP) failure potential and risk estimate changes. The effect of inspections has also been taken into account. The main tools for the pipe break probability and risk analyses have been probabilistic VTTBESIT code and Markov process application, but some calculations were also performed using Swedish NURBIT code for comparison. In the analyses, the considered degradation mechanism was stress corrosion cracking (SCC). During the three previous years, four representative NPP piping welds were considered. The computational part of this report concerns one of them. The initial flaw sizes have a considerable effect on the failure probability results. Selected distributions for the fabrication induced cracks provided the initial flaw state for all analysis cases, whereas for the cases with SCC induced initial cracks developed by VTT and with those in the NURBIT code, SCC flaws initiated during operation were used as well. The break probabilities after one year in operation are the smallest for the cases with fabrication induced cracks alone, whereas these probabilities are from 0.5 to 2 decades higher for the cases with the SCC induced initial cracks developed by VTT and with those in the NURBIT code. The effect of the initial flaw sizes to break probabilities is most pronounced in the early phase of the time in operation. Whereas the maximum break probability values after 60 years in operation are almost matching for all cases. The SCC induced initial crack sizes developed earlier in this study are recommended to be used. The magnitude of the loading has the largest effect on the break probability results. The loading is dominated by the weld residual stresses (WRS). WRSs from several different sources were included in the study and differences were large with regard to break probability. For welds joining NPP pipes of austenitic stainless steel, the WRSs given in the SSM handbook and SINTAP procedure are recommended to be used.