Effect of Analytical Methodology for Assessment on Seismically Induced Core Damage

Abstract

The purpose of this study is to investigate the effects of the difference between analytical methodologies for seismic probabilistic safety assessment (PSA), especially the difference in the probability of system failures and core damage sequences. We selected two methods, upper bound approximation (UBA) using minimal cut sets and direct quantification using Monte Carlo simulation (DQFM) for comparison. The former is often used in PSA for both internal and external event. The latter is developed in Japan Atomic Energy Agency (JAEA) as a part of SECOM2, a system reliability analysis code for seismic PSA, and expanded to divide the core damage state into different core damage sequences. The result showed that there is a significant difference between UBA method and DQFM method.

Efficiency of Analytical Methodologies in Uncertainty Analysis of Seismic Core Damage Frequency

“…On the other hand, it was observed that the approximation error of MCS method in point estimation can cause CDFs per accident sequence to be inconsistent in the similar degree with those presented in Ref. (8); while some accident sequences were overestimated, others were underestimated, and hence the rank of dominant accident sequences can be misleading. However, the result of this paper has shown that the inconsistency in using the MCS method abated in the uncertainty analysis since the right tails of overestimated CCDP distributions are limited by the probabilistic upper bound of 1.0.…”

“…Given that objective, a hypothetical BWR plant PSA model 22 is utilized in this study because the model has been well examined and documented 8,9,12 . This model consists of five initiating events and mitigation systems to ease accident progression induced by the initiating events.…”

“…This is because (1) in MCS method only approximates the CDF, (2) the accuracy of the approximation gets worse as the probability of each term in the cutset in computation becomes higher, and (3) the probability of each term in the cutsets are actually high in seismic CDF calculations in general. This problem has been recognized in the literature and hence the effects of the approximation on the mean value of CDF have been examined [8][9][10][11] .…”

Effect of Simultaneous Consideration for Seismically Induced Events on Core Damage Frequency

“…The results also show that it is important to choose the appropriate method among the analytical methods available even in the integrated model, and the impact on the CDF is significant as shown in the former study 13 .…”