Vessel rupture: Synthesis of the FOREVER interpretation activity

At FZR an integral axisymmetric finite element model has been developed for simulat-ing the late phase of core melt down scenarios in a RPV. The model allows for the calculation of the failure time and the failure mode of a vessel with a heated melt pool. In the thermal submodel the transient temperature field of the melt and of the vessel wall is evaluated. This can be done either with a CFD model or with the Effective Conduction Convection Model (ECCM). Within the mechanical submodel the viscoplastic deformation of the vessel wall is simulated. By use of the material damage the failure time and position can be deter-mined. An additional mechanical submodel is used to evaluate the melt pool deformation. The thermal and the mechanical submodels are recursively and sequentially coupled, i.e. for each time step a thermal dynamic and a mechanical solution is calculated considering a mutual feedback. Besides of the temperature dependence of the material parameters and the thermally induced stresses, which are considered in the mechanical model, also the consequences of the vessel deformation for the temperature calculation are included (change of melt pool geometry, melt level drop, change of heat resistance of the vessel wall through thickness reduction, increase of the effective surface for heat radiation and convection). For the mechanical model a creep data base has been developed based on experimental results of projects of the 4th frame work programme of the EU. The viscoplastic calculation is coupled with the material damage. The creep data base was validated using two pipe rupture experiments. The coupled model for vessel failure was for pre- and post test calculations of the FOREVER experiments performed at the KTH Stockholm. In general, a good agreement of the calculation and the experimental results could be achieved. The FE model can be considered as validated for medium scaled vessel tests. Some sensitivity studies are done to evaluate the influence of the design peculi-arities of the FOREVER experiments in comparison with the prototypic scenarios.