During a severe accident in a nuclear power plant, hydrogen can be generated, leading to risks of possible deflagration and containment integrity failure. To manage severe accidents, great experimental, analytical, and benchmarking efforts are being made to understand combustible gas distribution, deflagration, and detonation processes. In one of the benchmarks—SARNET H2—flame acceleration due to obstacle-induced turbulence was investigated in the ENACCEF facility. The turbulent combustion problem is overly complex because it involves coupling between fluid dynamics, mass/heat transfer, and chemistry. There are still unknowns in understanding the mechanisms of turbulent flame propagation, therefore many methods in interpreting combustion and turbulent speed are present. Based on SARNET H2 benchmark results, a two-dimensional computational fluid dynamics simulation of turbulent hydrogen flame propagation in the ENACCEF facility was performed. Four combustible mixtures with different diluents concentrations were considered—13% H2 and 0%/10%/20%/30% of diluents in air. The aim of this numerical simulation was to validate the custom-built turbulent combustion OpenFOAM solver based on the progress variable model—flameFoam. Furthermore, another objective was to perform parametric analysis in relation to turbulent speed correlations and turbulence models and interpret the k-! SST model blending function F1 behavior during the combustion process. The obtained results show that in the simulated case all three turbulent speed correlations behave similarly and can be used to reproduce observable flame speed; also, the k-" model provides more accurate results than the k-! SST turbulence model. It is shown in the paper that the k-! SST model misinterprets the sudden parameter gradients resulting from turbulent combustion.
Loading....