Human event observations in the individual plant examinations

摘要

A major objective of the Nuclear Regulatory Commission's (NRC) Individual Plant Examination (IPE) Insights Program is to identify the important determinants of core damage frequency (CDF) for the different reactor and containment types and plant designs as indicated in the IPEs. The human reliability analysis (HRA) is a critical component of the probabilistic risk assessments (PRAS) which were done for the IPES. The determination and selection of human actions for incorporation into the event and fault tree models and the quantification of their failure probabilities can have an important impact on the resulting estimates of CDF and risk. Therefore, two important goals of the NRCs IPE Insights Program are (1) to determine the extent to which human actions and their corresponding failure probabilities influenced the results of the IPEs and (2) to identify which factors played significant roles in determining the differences and similarities in the results of the HRA analyses across the different plants. To obtain the relevant information, the NRC's IPE database, which contains information on plant design, CDF, and containment performance obtained from the IPES, was used in conjunction with a systematic examination of the HRA analyses and results from the IPES. Regarding the extent to which the results of the HRA analyses were significant contributors to the plants' CDFs, examinations of several different measures indicated that while individual human actions could have important influences on CDF for particular initiators, the HRA results did not appear to be the most significant driver of plant risk (CDF). Another finding was that while there were relatively wide variations in the calculated human error probabilities (HEPs) for similar events across plants, there was no evidence for any systematic variation as a function of the HRA methods used in the analyses.