Assessment of CFD hydrogen deflagration models for application in light water cooled nuclear reactor containments

Large quantities of hydrogen can be generated during a severe accident in a water-cooled nuclear reactor due to degradation of the core. Hydrogen is a flammable gas that can form a combustible mixture with air. When released in the air-filled containment of a pressurized water reactor, the hydrogen can thus create a potential deflagration risk. The dynamic pressure loads resulting from hydrogen combustion can be detrimental to the structural integrity of the reactor safety systems and the reactor containment, hence pose a great threat to the environment. Therefore, accurate prediction of these pressure loads is an important safety issue. During a severe accident, cooling water from the primary circuit will enter the containment as steam. Therefore, it is important to capture the effect of steam on hydrogen deflagration in a CFD model. This thesis work follows from a larger framework research concerning the hydrogen deflagration issue developed at the Nuclear Research and Consultancy Group (NRG) in The Netherlands. In past NRG assignments, a Computational Fluid Dynamics (CFD) based method has been validated to determine the pressure loads from a fast deflagration for uniform hydrogen-air mixtures, hydrogen-air mixtures with diluents and non-uniform hydrogen-air mixtures. The combustion model applied in the CFD method is based on the Turbulent Flame Speed Closure (TFC) of Zimont and is implemented in the CFD software ANSYS Fluent using user defined functions. In a more recent step, the extension of the above combustion model, Extended Turbulent Flame Speed Closure (ETFC) of Lipatnikov, which includes the laminar source term, has been studied, and its validation against hydrogen deflagration experiments in the slow deflagration regime for uniform hydrogen-air mixtures has been evaluated. The primary objectives of the present work are to further validate the TFC and ETFC combustion models in the slow deflagration regime for uniform hydrogen-air-steam mixtures, and investigate their capability to predict the effect of steam on hydrogen deflagration. Experiments conducted in a medium-scale Thermal-Hydraulics Hydrogen Aerosols and Iodine (THAI) test facility are used for validation purpose. In particular, three THAI hydrogen deflagration (HD) experiments with vertical flame propagation and increasing concentration of steam are considered, namely, THAI HD-15, HD-22 and HD-24. This thesis work describes the HD experiments from the medium scale THAI facility, the applied combustion models, and the most relevant results of the code validation. The peak pressures, the trends of the flame velocity, and the pressure rise with an increase in the initial steam dilution are captured reasonably well by both combustion models. In addition, the ETFC model appeared to be more robust to mesh resolution changes. The mean pressure rise is evaluated with 18% under-prediction and the peak pressure is evaluated with 5% accuracy, when steam is involved