Curvature effects in laminar and turbulent non-premixed combustion

点击翻译按钮获取中文摘要

In funding period 1 of the project, a comprehensive understanding of curvature effects in non-premixed combustion was achieved. In 10 peer-reviewed publications, it was shown that curvature-induced effects significantly influence the microstructure of laminar and turbulent diffusion flames. In addition to the simulation of canonical laminar flames, direct numerical simulations (DNS) of time-evolving turbulent jet flames (H2-air, syngas-air) were performed for this purpose, thus establishing a comprehensive reference database for the analysis of curvature effects. The analyses were summarized in a regime diagram for the assessment of curvature effects. A methodological highlight of funding period 1 is the in situ tracking of gradient trajectories (GTs) of the mixture fracture field in the DNS; individual GTs or flamelets can be tracked and analyzed in their spatial and temporal evolution. These Lagrangian data can be used to evaluate central model assumptions of the flamelet concept in turbulent flames. It has been shown that consideration of curvature in the flamelet equations alone is not sufficient for the prediction of curvature effects. Curvature-induced tangential diffusion along mixture fracture isosurfaces is the dominant effect in the case of large curvatures, which can only be fully described by the extended Flamelet equations. In the second funding period, the curvature influence on the auto-ignition of non-premixed turbulent flames will be investigated. In contrast to a (statistically) stationary diffusion flame, this is the preceding, highly transient process in which an unreacted (or slowly reacting) mixing layer quickly develops into a diffusion flame after an ignition delay time. Experiments and direct numerical simulations in the literature demonstrate that ignition kernels form along isosurfaces of a preferred mixture fraction, the "most reactive mixture fraction," in pockets of low scalar dissipation rate and negative curvature. Curvature effects can thus significantly influence ignition, but previous flamelet approaches have so far focused almost exclusively on the influence of scalar dissipation rate. The extension of flamelet theory to include curvature effects for steady-state conditions from funding period 1 is the starting point for modeling the transient auto-ignition of non-premixed flames. In addition to the analysis of the DNS data and the flamelet development, the transfer to the LES in funding period 2 is also targeted.