Simulation of the core degradation phase of the Fukushima accidents using the ASTEC code

The French Institute for Nuclear Safety and Radioprotection (IRSN) attempts to simulate the Fukushima accidents using the ASTEC integral code. This paper summarizes the main results of the simulations conducted before the beginning of the OECD/NEA/CSNI Benchmark Study of the Accident at the Fukushima Daiichi Nuclear Power Station (BSAF) project. rst analysis carried out concerned the unit 2 transient. Results were considered as satisfactory being quite consistent with measures reported by TEPCO and similar computations performed with MELCOR or MAAP. Knowledge gained from PWR practice and different lectures available in the open literature for BWR provided valuable technical elements to explain observations or to validate assumptions. Leakage model from the containment up to the refuelling bay through the head flange seal was very efficient to retrieve pressure evolution inside the dry well. Extension of the model to reactor number 3 gave also results quite consistent with what similar codes computed. However for both reactors some figures characteristic of the transient as hydrogen production are liable to vary a lot if models for bottom and top nozzles are added which has not been done in reference computation due to present lack of data. Uncertainties with simulation of accident on reactor number 1 are rather large due to the scarcity of data. Further, as the measurement points were quasi absent for most of the first 24 h there is no reference to compare to simulation results. Bottom vessel head failure is predicted but due to the high number of penetrations the mechanical failure models developed for PWR may not be so relevant for BWR.