Investigation of Flammability of Hydrogen Gases With Diluent Gases Under Severe Accident Conditions Using CNFT Model

Abstract

After the Fukushima Daiichi accident, predicting lower flammability limits (LFL) as a part of hydrogen risk analysis has become an ever important task. Although many experimental studies have been conducted extensively, the LFL results for mixtures abided by the severe accident conditions are still lacking. The objective of this study is to develop a calculated non-adiabatic flame temperature (CNFT) model, which facilitates to predict the LFL of hydrogen mixtures. This model considers heat loss due to radiative heat transfer from flame to ambient environment during flame propagation. The model shows better agreement with experimental results for various mixtures than previous model, which predicts the LFL through a calculated adiabatic flame temperature. Especially, prediction accuracy for H2-air-steam mixture and mixtures at elevated initial temperature is improved substantially. Thus it is worth to evaluate the applicability of the CNFT model in the hydrogen risk analysis during severe accident. The postulated hydrogen risk in the current Optimized Power Reactor 1000 MWe (OPR1000) under Station Blackout (SBO) scenario was investigated with MELCOR 1.8.6 code. As a result, it was observed that uncertainty of hydrogen risk calculated with the MELCOR default model can be reduced by the CNFT model. This study suggests that the developed CNFT model can enhance reliability of severe accident analysis related to the flammability of hydrogen mixtures.