The University of New South Wales

 

Significant improvements of the fuel efficiency and reductions of emissions (including CO2) from combustion equipment are required to develop the next generation of clean energy technologies. Impeding this progress is a basic lack of understanding of the multi-scale, multi-physics turbulent flows that underpin the performance of these technologies. Direct numerical simulations have long been recognized as a key method to obtain this lacking understanding but until recently these methods were limited to very idealized scenarios. However, recent major advances in supercomputing hardware have enabled rapid progress, such that DNSs of laboratory-scale experiments are now possible given sufficient allocations of computer time. In this talk I will discuss two examples recently performed by our group at the University of New South Wales.

 

In the first example, I discuss DNS of lifted, slot-jet methane-air flames having a parameter regime comparable to laboratory experiments. The stabilization mechanism is analyzed in terms of the propagation velocities of edge-flames at the flame base. It is shown that, on average, the flow velocity balances the relative displacement speed of the edge-flames, thus demonstrating that the flame is stabilized by edge-flame propagation. Fluctuations of the edge-flame locations, however, are shown to be quite significant. Further analysis shows that edge-flames tend to move in a cyclic rotating pattern, which I argue is due to the passage of large eddies.

 

In the second example, I discuss DNS of an experimental high Karlovitz number, lean, premixed methane-air jet flame using a reduced 28 species chemical mechanism. Good agreement is revealed in comparisons between DNS and experiments are revealed, both for quantitatively for temperature and velocity, and qualitatively for planar laser-induced fluorescence of key species. Analysis of the results reveals significant mixing resulting in flame broadening both within the preheat zone as well as in the reaction zone, supporting the experimental observations.

 

Evatt Hawkes is an Australian Research Council Future Fellow and Associate Professor at the University of New South Wales. Prior to UNSW, he received his PhD from the University of Cambridge in 2001, and then worked as a post-doctoral fellow at the Combustion Research Facility of Sandia National Laboratories, USA. At UNSW, he leads a program of computational and experimental energy research focusing on internal combustion engines and solar thermochemical energy systems. In the engine-combustion area, the group’s work spans applied studies of low-emissions and alternatively fueled engines to fundamental investigations of combustion using high fidelity numerical simulations and using laser-diagnostics in optically accessible research engines. In particular, Hawkes is known for ground-breaking work with direct numerical simulations executed using massively parallel supercomputing resources.

联系人：杨越  62745328