Global sensitivity and uncertainty analysis on the neutronics study of the Chinese Experimental Fast Reactor (CEFR)

The aim of this benchmark study is to perform the Sensitivity and Uncertainty Analysis on the neutronics analysis of the China Experimental Fast Reactor (CEFR), which corresponds to the last work package of the ongoing IAEA Coordinated Research Project based on CEFR start-up tests. The sensitivity analysis was performed using the brute force method for single and combined parameters as well as for the neutron cross-sections based on the ENDF/B-VII.1 nuclear data library. The most sensitive input parameter was found to be the mass of U235 followed by fuel rod diameter, which increased the critical effective multiplication factor of the system by 21452 and 6023 pcm when a positive increase of 10% and 5% incurred from their nominal values, respectively. On the other hand, the most sensitive nuclear reaction corresponded to fission of U235. Regarding the uncertainty analysis, only input parameters related to neutron induced reactions of selected nuclides were considered as sources of uncertainty and, based on a statistical methodology, these were sampled using the so-called SANDY nuclear data processing code. Thus, the Total Monte Carlo (TMC) approach was employed for the uncertainty propagation through the OpenMC code. For the uncertainty assessment in observables of interest, the Wilks’ distribution free method was applied to infer (at least) the first order 95%/95% tolerance limits of the effective neutron multiplication factor while employing both the ENDF/B-VII.1 and JEFF3.3 libraries. After 300 calculations, an estimation of the positive limit on the effective multiplication factor samples based on ENDF/B-VII.1 and JEFF3.3 turned to be 1.05947 and 1.03361, respectively. Thus, by following the same conclusion, the relative standard deviation for the two libraries corresponded to be of 2.455% and 1.611%, respectively.