HLRB Project pr53va

Numerical Modelling of Turbulent Boundary Layer Flashback

Lehrstuhl für Thermodynamik,TUM

Numerical Modelling of Turbulent Boundary Layer Flashback

Lehrstuhl für Thermodynamik,TUM

Lehrstuhl für Thermodynamik,TUM

Aaron Ömer Endres

Boltzmannstraße 15

85747 Garching

Flame flashback from the combustion chamber into the upstream burner mixing section poses an inherent threat associated with premixed combustion. Especially for hydrogen-rich fuels boundary layer flashback (BLF) is likely to occur and damage the combustor. It is thus important to develop methods for the accurate prediction of the onset of BLF.In this project BLF is to be modelled numerically within the framework of computational fluid dynamics with OpenFOAM. So far, the numerical prediction of the onset of turbulent BLF of hydrogen flames could not be achieved. Therefore, the ability of Large Eddy Simulations (LES) to reproduce experimentally determined flashback limits of hydrogen flames is investigated. First, inert LES of the turbulent channel flow are conducted. The effect of the chosen turbulence model on the accurate prediction of the turbulence characteristics is investigated. The unsteady velocity profile gained from the inert simulations can then be used as an inlet boundary condition for reactive LES. The combustion process in the reactive LES is modeled by solving a transport equation for each species. The chemical source term is calculated by solving a detailed chemical kinetics mechanism. Within this approach the influence of the subgrid turbulence-chemistry interaction is neglected. By accurately modelling the species and heat diffusion processes and with a high flame front resolution the propagation of the flame is still expected to be accurately predicted.Besides the BLF limits and the influence of modelling parameters on the flashback limits, a detailed insight into the process of BLF can be obtained from the simulations. Especially the transition of a flame stabilized at the exit of the channel to a propagating flame is not yet fully understood. With the transient characteristics of the LES and the high spatial resolution, the remaining open questions for BLF are expected to be clarified. Furthermore, the data gained from the numerical simulations can be used for developing analytical prediction methods for BLF.

Impressum, Conny Wendler