![Sample our Physical Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110819/TOC-Subject-Sample-2021-Physical-Sciences.jpg"})
Abstract
In the framework of the LWR-PROTEUS programme, integral experiments are being performed for the validation of reactor physics methods/data applied to the analysis of modem light water reactor (LWR) fuel assembly designs. One of the important aspects pertaining to this programme is the investigation of the sensitivity of integral quantities, such as reaction rate spatial distributions, local reaction rate ratios and reactivity effects, to different nuclear data sets, e.g. ENDF/B-V, ENDF/B-VI and JEF-2.2. Data sensitivity studies are currently reported, largely based on continuous-energy Monte Carlo calculations at PROTEUS-core, BWR fuel assembly and pin-cell levels. Data effects could thus be investigated systematically, independent of the inherent limitations of deterministic codes such as geometrical simplification and resonance shielding approximations. The results show a significant sensitivity of system reactivity to the data set employed, calculated values being larger by several hundreds of pcm with JEF-2.2 than with ENDF/B-V or B-VI. As regards detailed reaction rate distributions in BWR assemblies (investigated in LWR-PROTEUS Phase I), there appears to be only a weak dependence on the data library used. In the context of the investigations planned with highly burnt fuel rod segments in the Phase II experiments, certain studies made at pincell level have indicated that the effects of basic data differences are likely to be much lower than uncertainties in fuel composition predictions.