Quantifying uncertainty induced by scattering angle distribution using maximum entropy method

Maruyama, Shuhei   ; Yamamoto, Akio*; Endo, Tomohiro*

This study developed a new method for evaluating the uncertainty in reactor core/shielding characteristics attributable to the scattering angle distribution, employing a random sampling (RS) technique integrated with continuous energy Monte Carlo (CEMC) calculations. The impact of neutron scattering angle is not negligible in the analysis of fast reactor cores and shielding. Recent advancements have enabled the high-accuracy assessment of nuclear data-induced uncertainty by merging CEMC calculations and the RS technique. Nonetheless, a method to quantify uncertainty due to scattering angle distribution remains unestablished. This study introduces a new approach for uncertainty quantification related to scattering angle distribution in CEMC-RS, utilizing the maximum entropy method. The effectiveness of this method was verified through comparison with results from the classical deterministic uncertainty quantification approach based on generalized perturbation theory. Overall, this method offers a more accurate tool for nuclear engineers and researchers in evaluating and managing uncertainties in reactor design and safety analysis.