Ability of Current Advanced Codes to Predict Core Degradation, Melt Progression and Reflooding

The experimental database on core degradation and melt relocation (and their consequences on hydrogen production, vessel rupture) is limited to small-scale experiments which are only partially representative of what could occur in a reactor. As a consequence, there is uncertainty in the capability of codes to predict core degradation in postulated severe accident transients of nuclear power plants. The GAMA has launched an action in order to determine the ability of current advanced codes to predict core degradation in nuclear reactors. The TMI-2 scenario was selected as the case to analyze since it concerns the only full scale Pressurized Water Reactor to have experienced core degradation. Data from the code calculations were compared to the TMI-2 end-state to determine the codes‟ predictive capability. The study was completed in 2004, and is documented in ref. 2. One conclusion of the study is that variability in the codes predictions existed in part because initial conditions of the tMI-2 scenario were not well defined. It was concluded that code variability could be better evaluated if these conditions were better defined. Therefore, an additional task was proposed to benchmark the codes. This phase II study evaluates the variability in the codes‟ results using a postulated core degradation scenario of the TMI-2 reactor. The scenario was specified with simple initial and boundary conditions so that the influence of uncertainty of these conditions was minimized and the variability in the codes‟ results is more readily determined