SANDA Project D5.6: Report on correlation between integral experiments

Abstract

The main purpose of SANDA project Task 5.2 is to assess the JEFF evaluated nuclear files using existing experiments, with the aim to contribute to the development of the future JEFF-4.0 nuclear data library. Adjustment and assimilation studies are powerful techniques to make the most of the information provided by experimental measurements and reducing biases in the calculation results and providing recommendations for nuclear data improvement. These techniques usually need to employ a large set of integral experiments to provide consistent information. However, the parameters of the experiments within this set can be correlated. Therefore, for the application of adjustment and assimilation techniques it is essential to have some prior knowledge of these correlations. For this reason, subtask 5.2.1, entitled “Assessing correlations in integral experiments” has been included within SANDA project Task 5.2. This report presents the results of this subtask. It includes a survey of the available correlations between integral experiments in the ICSBEB criticality experiment database, a review of the status of correlations between shielding benchmarks and some applications to particular cases, namely: (1) two cores of the EOLE facility investigated under the CAMELEON program; (2) a set of six cores loaded in the ZPR facility considered to be of interest for validation of nuclear data for Sodium Fast Reactors (SFRs); (3) reaction rates measured at the ASPIS Iron-88 benchmark experiment, and (4) an example of the application of Machine Learning techniques in the interpretation of uncertainties and correlations in a benchmark experiment.