혼합효과 모델을 이용한 원자로 압력용기 조사취화 경향곡선의 실용적 조정

Practical Adjustment of Embrittlement Trend Curves for Reactor Pressure Vessels Using a Mixed-Effect Model

This study proposes a practical adjustment equation for the Embrittlement Trend Curve (ETC) to effectively apply it to reactor pressure vessel (RPV) materials in individual nuclear power plants. Traditional ETC adjustment methods have limitations due to constraints in number of group-specific measurements and a lack of statistical foundations. To address these issues, KAERI applied a Markov Chain Monte Carlo (MCMC)-based mixed-effect model to the latest ETC model, ASTM E900-15. This approach quantitatively calculates the mean, standard deviation, and prediction intervals of the adjustment intercept by considering the grouping characteristics of surveillance data and uncertainties in unirradiated specimens. Although the KAERI model provides quantitative distributions of parameters and intercepts, it has challenges in practical applications due to computational complexity and low portability. In this study, a simplified equation was developed using the statistical calculations of the mixed-effect model, which retains the primary outcomes of the KAERI model while enhancing portability. This equation supports effective adjustments to the ASTM E900-15 ETC for nuclear power plants with diverse material properties and operational conditions. It enables reliable evaluations of RPV integrity using plant-specific surveillance data. The findings of this study are expected to improve the precision and practicality of ETCs.