Fuzzy safety analysis of LOCA accident in a heavy water research reactor

Safety analysis of nuclear reactors is one of the most important procedures as part of licensing process. Probabilistic safety assessment (PSA) which plays an undeniable role in risk evaluation provides insights into design weaknesses and into ways of preventing core damage, which in most cases is the precursor to accidents leading to major radioactive releases with potential health and environmental consequences. Despite PSA merits, some shortcomings associated with the final results exist. Conventional PSA uses crisp values to represent the failure probabilities of basic events which will cause a high level of uncertainty due to the inherent imprecision of input data. In this study, to cope with this problem, a fuzzy approach has been employed. Through this method, the basic events crisp values are replaced with fuzzy numbers as failure probabilities. Hence, in the fault tree and event tree analysis, the top events and the end-states frequencies are treated as fuzzy numbers. Moreover, in order to identify the components and safety systems which are critical and have the maximum contribution to the uncertainty, fuzzy importance measures and fuzzy uncertainty index are developed.