Development of extreme rainfall PRA methodology for sodium-cooled fast reactor

ナトリウム冷却高速炉に対する異常降雨PRA手法の開発

西野 裕之  ; 栗坂 健一 ; 山野 秀将   

Nishino, Hiroyuki; Kurisaka, Kenichi; Yamano, Hidemasa

The objective of this study is to develop a probabilistic risk assessment (PRA) methodology for extreme rainfall with focusing on decay heat removal system of a sodium-cooled fast reactor. For the extreme rainfall, annual excess probability depending on the hazard intensity was statistically estimated based on meteorological data. To identify core damage sequence, event trees were developed by assuming scenarios that structures, systems and components (SSCs) important to safety are flooded with rainwater coming into the buildings through gaps in the doors and the SSCs fail when the level of rainwater on the ground or on the roof of the building becomes higher than thresholds of doors on first floor or on the roof during the rainfall. To estimate the failure probability of the SSCs, the level of water rise was estimated by comparing the difference between precipitation and drainage capacity. By combining annual excess probability and the failure probability of SSCs, the event trees led to quantification of core damage frequency, and therefore the PRA methodology for rainfall was developed.