Sensitivity Analysis for beyond Design Basis Accidents of VVER-1200 Power Unit
摘要
In accordance with the current requirements of Russian and foreign regulatory authorities, the modeling of beyond design basis accidents must be accompanied by a sensitivity analysis of the results of calculation. The outcome of the sensitivity analysis is the determination of the degree of influence of each uncertainty on the parameters under investigation. Within the scope of the performed work, an approach to performing sensitivity analysis using the SUSA program and the TRAP-KS software package was proposed, demonstrated by the example of beyond design basis accidents such as “trip of different number of reactor coolant pump sets with common-cause failure of programmable automatics systems,” “loss of vacuum in the condenser or other cases leading to turbine trip with common-cause failure of programmable automatics systems,” and “loss of nonemergency a.c. power to the station auxiliaries (loss of NPP power supply) with common-cause failure of programmable automatics systems.” These scenarios were selected on the basis of the condition of achieving maximum values of acceptance criteria: maximum cladding temperature, maximum pressure in the primary and secondary circuits, respectively. The results of the sensitivity analysis presented in this work include values characterizing the spread of the criterion parameters (maximum cladding temperature, maximum pressure in the primary and secondary circuits), Spearman rank correlation coefficients, and probability density function graphs and distribution charts of the results of calculation. The approach discussed in this paper allows for compliance with the current requirements of international and Russian supervisory bodies and can later be extended to all NPP projects with VVER reactor units. Additionally, this approach may be further applied to beyond design basis accidents involving failures of the accident protection system (ATWS) and failures of the industrial circuit due to a common cause.