Analysis of LBLOCA using best estimate plus uncertainties for three-loop nuclear power plant power uprate

The best estimate (BE) calculation with uncertainty evaluation of large break loss of coolant accident (LBLOCA) has been increasingly applied to the licensing applications of nuclear power plants (NPPs). The KINS-realistic evaluation methodology (KINS-REM) was developed for the independent audit calculation based on the best estimate plus uncertainty (BEPU) method, and the code accuracy and statistical method have been improved. Power uprate has been implemented at a number of NPPs in many countries. It is generally categorized based on the increment of power and the method of increasing power. In this study, assuming the 4.5% stretch power uprate of typical three-loop nuclear power plant and corresponding design modifications, the LBLOCA was analyzed by applying the KINS-REM. The MARS-KS code was used as a frozen BE code, and 18 uncertainty parameters were considered in the analysis. The nodalization of the three-loop plant was defined by implementing specific separate and integral effect test nodalization scheme, and the thermal-hydraulic behavior in the system during the LBLOCA was analyzed through the basecase calculation. The 95 percentile peak cladding temperature (PCT) with 95% confidence level was determined by 124 calculations based on Wilks' formula of the non-parametric statistics, and additional biases for emergency core coolant (ECC) bypass and steam binding were added to account for untreatable phenomena and models. It was confirmed that the analysis results of the LBLOCA for the three-loop plant power uprate met the PCT acceptance criteria.