Speaker
Description
The China Experimental Fast Reactor (CEFR) is a 65 MWth pool-type sodium cooled fast reactor. The International Atomic Energy Agency (IAEA) launched a Coordinated Research Project (CRP) based on different released and available start-up tests of CEFR when it first reached criticality in 2010. Therefore, the main objectives of this CRP aim at improving fast reactor simulation analytical capabilities, international validation and qualification of codes currently employed in the field of fast reactor neutronics.
The benchmarking of the CEFR start-up tests has been divided by the CRP coordinators in different work packages (WP), being the last one (WP8) devoted to perform sensitivity and uncertainty analysis into the different CRP neutronics studies. So far the first study, which corresponds to benchmarking the core fuel loading steps that are required to reach criticality, has been agreed by the CRP members to be used as a study case for testing the effect that different evaluated covariance data have on integral observables such as the effective neutron multiplication factor of the domain. Thus, an uncertainty analysis is expected to be performed to quantify how far the uncertainty associated to nuclear data may drive the CEFR neutronics model sub- or super-critical. For this purpose, a statistical approach has been adopted, where neutron microscopic cross-sections will be considered as random variables and sampled according to the recorded covariance information contained in different major nuclear data libraries such as ENDF/B-VII.1, JEFF-3.3, ENDF/B-VIII.0 and JENDL-4. In the end, microscopic cross-sections of important and selected nuclides will be considered as the only source of input uncertain parameters, where 300 random files of nuclear data (for each major library) will be created by means of the SANDY code. Thereafter, the propagation of such sources of uncertainty will be done by executing an eigenvalue transport calculation of the CEFR model previously created in the Monte Carlo-based OpenMC code. Thus, the uncertainty quantification on the criticality level of the model (done for each major nuclear data library) will follow a non-parametric statistical analysis based on a sample formed by 300 effective multiplication factor calculations. With this sample size the 5th and 95th percentile of the observable population of interest can be inferred with at least a 95% confidence (i.e. Wilk’s formulation).
The objective of this work is clear; to perform a comparison on how different covariance evaluations of important nuclear data libraries impact computed neutronic integral observables of fast reactors. Therefore, the organized IAEA-CEFR-CRP presents a unique opportunity to perform code validation in a holistic way, where the effect that nuclear data uncertainty coming from major evaluations around the world can be compared on fast reactor applications.