Statistical uncertainty analysis applied to fuel depletion calculations

In this paper, we present an uncertainty methodology based on a statistical approach, for assessing uncertainties in lattice code predictions of fuel composition changes with burnup due to uncertainties in the fuel geometric configuration, initial enrichment, and depletion conditions. The methodology has been applied to depletion calculations with CASMO-4 and experimental data from the ARIANE Programme to estimate the calculation uncertainties in nuclide concentration and other neutronic parameters at any time during the irradiation history. Results have shown that important information on the quality of the code’s predictions can be obtained by analyzing the comparison of the code’s estimates and their associated uncertainties, in the form of tolerance intervals, with experimental data and their reported errors.