Generation of fission yield covariances to correct discrepancies in the nuclear data libraries

Fission yield uncertainties and correlations should be considered in the uncertainty quantification of burnup responses — e.g. isotopic inventory, effective neutron multiplication factor keff . Although nuclear data libraries generally provide independent fission yield uncertainties along with the best estimates, currently they lack complete covariance matrices. In addition, several inconsistencies were detected amongst the current fission yield evaluated uncertainties, which could impact on uncertainty quantification (UQ) studies. As a part of this work, we introduced fission yield correlations to sort out the data
inconsistency found in the JEFF-3.1.1 fission yield library. Such correlations are produced using an iterative generalised least square (GLS) updating technique, with conservation equations acting as fitting models. The process revises the fission yield estimates and covariances according to reliable evaluations,
when available, or conservation criteria. We chose to work with the PWR fuel rod model of the REBUS international program to test the new covariances, since experimental uncertainties on several concentrations are available. We propagated the original and updated fission yield covariances using a sampling approach and we quantified the uncertainty of keff and nuclide densities in the chosen burnup problem. The response uncertainty for keff and nuclide densities showed a sharp drop when using the new set of fission yield covariance matrices.