Validation of PSI best estimate plus uncertainty methodology against SPERT-III reactivity initiated accident experiments.

This research aims at validating the SIMULATE-3K code and complementing the results with the quantification of nuclear data uncertainties against the Special Power Excursion Reactor Test III (SPERT-III) experiments. To that aim, the SHARK-X methodology, under development at PSI, for the propagation of nuclear data uncertainties in CASMO5 2-D lattice calculations to 3-D core transient simulations is applied for the analysis of a SPERT-III super-prompt critical test conducted at cold startup conditions. Concerning transient results, both total power and reactivity show a good agreement with the measurements at the initial phase of power excursion, while a slight discrepancy is obtained at the final phase of the transient. The estimated uncertainties regarding both steady-state parameters such as k-eff and static reactivity worth, as well as dynamical quantities such as power pulse width and enthalpies are presented. Results show non-negligible sensitivity upon the employed nuclear data library. The uncertainty quantification results show relatively small biases for k-eff and reactivity. The uncertainty in peak power is around 3%, while it is negligible for the time to peak power and the pulse width. The time evolution of the standard deviation and skewness of the total power showed special shapes with relatively high maximum values. In addition, the uncertainty due to nuclear data in the two important safety parameters, i.e. maximum nodal fuel temperature and enthalpy reaches maximum value around 2% and 10%, respectively. [ABSTRACT FROM AUTHOR]