Comparison of MELCOR modeling techniques and effects of vessel water injection on a low-pressure, short-term, station blackout at the Grand Gulf Nuclear Station

Abstract

A fully qualified, best-estimate MELCOR deck has been prepared for the Grand Gulf Nuclear Station and has been run using MELCOR 1.8.3 (1.8 PN) for a low-pressure, short-term, station blackout severe accident. The same severe accident sequence has been run with the same MELCOR version for the same plant using the deck prepared during the NUREG-1150 study. A third run was also completed with the best-estimate deck but without the Lower Plenum Debris Bed (BH) Package to model the lower plenum. The results from the three runs have been compared, and substantial differences have been found. The timing of important events is shorter, and the calculated source terms are in most cases larger for the NUREG-1150 deck results. However, some of the source terms calculated by the NUREG-1150 deck are not conservative when compared to the best-estimate deck results. These results identified some deficiencies in the NUREG-1150 model of the Grand Gulf Nuclear Station. Injection recovery sequences have also been simulated by injecting water into the vessel after core relocation started. This marks the first use of the new BH Package of MELCOR to investigate the effects of water addition to a lower plenum debris bed. The calculated results indicate that vessel failure can be prevented by injecting water at a sufficiently early stage. No pressure spikes in the vessel were predicted during the water injection. The MELCOR code has proven to be a useful tool for severe accident management strategies.