Development of a dynamic event tree (DET) to analyze SBO accident in VVER-1000/V446 nuclear reactor

Abstract

Probabilistic safety assessment (PSA) is the main approach currently applied in safety assessment of nuclear power plants to calculate the core damage frequency (CDF). The conventional event tree/fault tree methodology, adopted in current PSA studies, has weaknesses such as nonrealistic modeling of operator and recovery actions, inability for functional reliability modelling of passive safety systems, using predetermined timing and order of accident sequences without any interaction of system responses, insensitivity of the system failure to uncertainties in the process physics and consequently its output is not comprehensive for safety considerations, risk management and decision making tasks. In this study to overcome these bottlenecks, a dynamic approach based on dynamic event tree (DET) is applied to calculate the CDF for the station blackout (SBO) accident in VVER-1000/V446 nuclear reactor. To implement the DET, deterministic and probabilistic models are developed using the plant data and the characteristics of the SBO accident. These models are then run simultaneously to investigate the time of operator actions, AC power supply recovery, and the availability of safety systems on the accident progression and consequences. The results show that the value of CDF calculated with DET is equal to 1.97e-6 (yr−1). The results also illustrates that the DET can provide useful insights concerning potential risk contributors and the effectiveness of operator and recovery actions.