Abstract

A hypothetical severe accident in a nuclear power plant has the potential for causing severe core damage, including a meltdown. To prevent or in the case of an already formed debris bed to limit the in-vessel core degradation, the basic severe accident management strategies consider the in-vessel reflooding to ensure the debris bed coolability.The purpose of our research was to understand the key processes and conditions related to the in-vessel debris bed coolability in the bottom reflooding conditions. Recently, experimental tests in the PEARL facility (IRSN, France) were performed to highlight the behaviour of the steam and water flow in a hot porous medium and to provide experimental data to validate 2-D and 3-D models for the debris bed reflooding. Our aim was to analyse chosen PEARL experiments performed at the atmospheric pressure. The objective was to analyse the importance of the uncertainties in the initial and boundary conditions on the simulation results and to assess the heat transfer modelling approaches. Simulations were performed using the MC3D code (IRSN, France).In general, the performed simulations are in good agreement with the experiments. The general features, in particular the water preferential entrainment in the bypass are recovered and the analysis of calculation gives further information on the mechanisms. In particular, the mechanism of water deviation in the bypass (2-D behaviour) is described. The hypothesis of water dragged by steam coming from the debris bed region cannot be supported. However, the simulation results are indicating a noticeable impact of the actual conditions as the water temperature and the initial support bed and bypass temperature. The simulations, varying the porosity of the test section, showed that this impact affects the flow configuration and is important for cases with the 2-D configuration. The reflooding capabilities in this configuration may depend strongly on the characteristics of the debris bed. Changes in the heat transfer modelling do not have greater effect on the simulation results.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call