Toward an improved methodology to simulate the initiating phase of a severe accident in a Sodium Fast Reactor

International audienceTo consolidate the safety assessment for Sodium Fast Reactors, hypothetical core disruptive accident sequences have been extensively studied over the past decades. These sequences are initiated by a very low probability coincidence of an accident precursor, such as the stop of primary and/or secondary pumps, and failure of one or more safety systems. Regarding Sodium Fast Reactor systems, a major concern is that the core, at nominal, is not necessarily in its neutronically most reactive condition. From the computational point of view, a severe accident scenario is generally broken down into subphases as the initiating phase, the transition phase, the expansion phase and the decay heat-removal phase for applying specific code to each phase. The present research project is dedicated to the analysis and the quantification of bias corresponding to the computational methodology employed to simulate the initiating phase of severe accidents on Sodium Fast Reactors. To evaluate these biases, an extension of existing modeling capabilities is required. To fulfill these needs, a new computational model has been developed. This new model is based on a multi-physics coupling and makes use of modern parallel computing architectures. The computational model has been applied to obtain new information in regards to neutronics and thermal-hydraulics models and their coupling