Compared with conservation evaluation model, best estimate plus uncertainty (BEPU) method can obtain more realistic results and gain larger license margins with respect to the safety criteria. In view of this, a BEPU method named 4S (SNERDI Statistical Solution for Safety) has been developed, according to the basic principles of evaluation model development and assessment of RG 1.203.
The characteristics of 4S method are as follows: The output uncertainty is quantified by using random sampling and propagation of input uncertainties. Global sensitivity analysis is used to support PIRT establishment. Uncertainties of model parameters are calibrated and validated by using separate effects tests considering measuring uncertainties. DAKOTA code is used for uncertainty and sensitivity analysis. An automatic BEPU analysis platform has been developed by coupling DAKOTA and different reactor safety analysis codes, and code calculations can be performed in parallel.
BEPU analysis of mass and energy release and containment pressure response of CAP1400 under a postulated double-ended cold leg break loss of coolant accident (DECL LOCA) has been carried out by coupling DAKOTA, a mass and energy release analysis code and a containment analysis code. In total, 21 uncertain input parameters are considered. To make the results more stable, the sample size is 124 and the third highest peak pressure is used as the pressure upper bound (with 95%/95% probability/confidence) based on Wilks’ formula. The calculated results show that the peak pressure upper bound is obviously lower than the present conservation method used in license application, with more than 10% analysis margin. Influences of input parameter uncertainties on the containment peak pressure have been analyzed, according to the partial rank correlation coefficients calculated by DAKOTA. The results show that the input parameters mainly affecting the peak pressure are the containment condensation heat transfer multiplier, initial containment temperature, break resistance, decay heat, initial containment pressure, Core Makeup Tank (CMT) resistance multiplier and initial containment humidity.