An improvement in RELAP5/MOD2 heat transfer calculation during reflood

Analyses of two large-break loss-of-coolant-accident (LOCA) experiments, namely, LOFT L2-5 and Semiscale S-06-3, were performed with RELAP5/MOD2/cy 36.04. Excessive cooling, which occurred right before final quench, has been found in both calculations. The causes of the excessive cooling may be quite complex during large-break LOCA calculation. Among all the possible reasons, improper modeling of the heat transfer calculation in the reflood heat transfer package has been identified as a large contributor to the discrepancy between calculated and measured cladding temperatures. The extension of the correlations to the conditions far beyond the applicable ranges of the correlations and the usage of a partition factor in determining total heat transfer coefficient are the two aspects that will be improved in this study. A flow-regime-based heat transfer modeling has been incorporated into the RELAP5/MOD2 reflood heat transfer package. A brief description of the original model, the revised model, and the results from these two models are presented in this paper