Development of methodology for evaluating severe accident management using SAM strategies and PSA

Abstract

After the Fukushima nuclear accident, countries that use nuclear power have been carrying out various follow-up measures to increase the safety of nuclear power plants in preparation for severe accidents. Domestic Nuclear Power Plants also are taking follow-up measures to prepare for severe accidents. As part of this, systems for maintaining the integrity of containmnet building and minimizing the release of radioactive materials into the environment are installed in nuclear power plants. However, the effect of the severe accident mitigation systems currently installed in domestic nuclear power plants is not evaluated quantitatively. Therefore, this study presents a method to evaluate the effect of a severe accident mitigation systems by using a severe accident management strategy and probabilistic methods. In order to use probabilistic methods for the evaluation of severe accident mitigation systems, the level 2 PSA structure was analyzed. After selecting the risk information to be used for the evaluation, the selected risk information was databased in accordance with the format of the level 2 PSA. Containment event trees are databased so that the containment failure probability, which is one of the risk information, can be calculated automatically in the step of containment event tree requantification. Thereafter, severe accident mitigation systems that requires evaluation are selected, and the decomposition event trees and containment event trees are modified to apply the selected severe accident mitigation systems. In addition, the severe accident management strategies and severe accident phenomena were analyzed to derive the combinations that could be used for the severe accident mitigation systems in each plant damage state, and re-quantification of containment event trees was carried out. In addition, the factors used to quantify the operator reliability were improved to evaluate how training enhancement and procedure improvement affect the maintenance of the containment building integrity. As a result of re-quantification of containment event trees based on the combination of plant damage states and severe accident mitigation systems, it was found that emergency core cooling system and containment spray system could not be operated at the same time. In addition, the most effective combination of severe accident mitigation systems in all severe accident situations was the case where EXECCS and CFVS were applied at the same time, which reduced about 18 % of the CFF analyzed in this study. Both EXECCS and CFVS were estimated to be effective in reducing the CFF by about 8 %, respectively. However, if only ERVC is operated, it is calculated to increase CFF. Therefore, it was judged that it is necessary to operate in situations where the containment depressurization means are secured. When operator reliability improved, it was calculated to reduce the CFF by up to about 1% at maximum, which showed that the hardware equipment is effective in mitigating major accidents. In case that the operator reliability is improved, it is calculated that the CFF is reduced by up to about 1 %, which shows that the improvement of human error is less effective than hardware improvement.