A method to estimate the initiating event frequency by Monte Carlo Simulation

Li Yuandan,Tong Jiejuan
2008-01-01
Abstract:Initiating event frequencies, as one of the most important set of data in Probabilistic Safety Assessment, are usually evaluated by following ways: 1) Statistic estimation of the historic events; e.g. Transient; 2) Experts judgment or the mechanism analysis, e.g. LOCA; 3) Specific system models (usually fault tree models), e.g. initiating events caused by the failure of support system. The way to evaluate initiating event frequency by fault tree method works theoretically, however, there are lots of difficulties in practical, such as dynamic switching among redundant trains, different maintenance and testing strategies and so on. This paper discusses the difficulties in using Fault Tree method to evaluate the initiating event probability. In order to account for these difficulties, a method based on Monte Carlo Simulation is proposed. Fundamental principles and processes of Monte Carlo Simulation used to simulate the system behaviors are introduced in the paper, and the specific issues during the application of Monte Carlo Simulation in initiating event frequency estimation are presented. A study based on a simplified sample seawater system in a nuclear power plant is also given. The feasibility of using Monte Carlo Simulation to estimate the initiating event frequencies can be basically approved by the study results.
What problem does this paper attempt to address?