Use of an influence diagram and fuzzy probability for evaluating accident management in a BWR

This paper develops a new approach for evaluating severe accident management strategies. At first, this approach considers accident management as a decision problem (i.e., ''applying a strategy'' vs. ''do nothing'') and uses influence diagrams. This approach introduces the concept of a ''fuzzy probability'' in the evaluation of an influence diagram. When fuzzy logic is applied, fuzzy probabilities in an influence diagram can be easily propagated to obtain results. In addition, the results obtained provide not only information similar to the classical approach using point-estimate values, but also additional information regarding the impact from imprecise input data. The proposed methodology is applied to the evaluation of the drywell flooding strategy for a long-term station blackout sequence in the Peach Bottom nuclear power plant. The results show that the drywell flooding strategy seems to be beneficial for preventing reactor vessel breach. It is also effective for reducing the probability of the containment failure for both liner melt-through and late overpressurization. Even though there exists uncertainty in the results, ''flooding'' is preferred to ''do nothing'' when evaluated in terms of expected consequences, i.e., early and late fatalities