Propagation of void fraction uncertainly measures in the RETRAN-3D simulation of the Peach Bottom turbine trip

The paper describes the propagation of void fraction uncertainty, as quantified by employing a novel methodology developed at PSI, in the RETRAN-3D simulation of the Peach Bottom turbine trip test. Since the transient considered is characterized by a strongly coupling between thermal-hydraulics and neutronics, the accuracy in the void fraction model has a very important influence on the prediction of the power history and, in particular, of the maximum power reached. It has been shown that the objective measures used for the void fraction uncertainty, based on the direct comparison between experimental and predicted values extracted from a database of appropriate separate-effect tests, provides power uncertainty bands that are narrower and more realistic than those based, for example, on expert opinion. The applicability of such an approach to NPP transient best estimate analysis has thus been demonstrated.