QUANTIFIED PIRT USING CHARACTERISTIC TIME RATIOS

摘要

The US Nuclear Regulatory Commission (NRC) revised reactor licensing rules in 1988 to allow the use of realistic best-estimate (BE) computer code if its uncertainties are quantified following the rigorous Code Scaling, Application and Uncertainty (CSAU) methodology. The CSAU methodology introduced Phenomena Identification and Ranking Table (PIRT) to identify important contributors to uncertainty. PIRT ranks the importance of uncertainty contributors based on expert judgments as well as calculations. In this study, Quantified PIRT (QPIRT) is proposed as an alternate approach to identify and rank the importance of physical phenomena. QPIRT is solely based on the ranking of numerical values of Π groups obtained from dimensionless analysis to field equations. The numerical values of Π groups are calculated using best-estimate computer codes (RELAP5) simulation results. Since Π groups of greater values have higher impact to the solution of field equation, QPIRT identify and prioritize important phenomena based on the ranking of associated Π groups. RELAP5 simulation to a reactor integral test facility is employed to demonstrate the QPIRT method. Besides, the obtained QPIRT is compared to an existing expert PIRT to identify the effectiveness and limitations of QPIRT The QPIRT, together with the expert PIRT, is recommended to be used to identify important phenomena and uncertainty contributors for next generation nuclear system computer code validation and uncertainty analysis.