Simulation of premixed turbulent flames

1. Introduction Premixed turbulent flames are of increasing practical importance and remain a significant research challenge in the combustion community. These flames have the potential to operate cleanly and efficiently over a broad range of fuels, and therefore to represent a key element in the implementation of low-emissions burners for a variety of industrial applications. However, it is difficult to design lean premixed systems that are both safe and reliable. Premixed systems require device-scale flame stabilization techniques to create a statistically stationary configuration, and at the same time operate in regimes where the dynamics of the inertial scales of the turbulence, and the interactions of the turbulence with the combustion chemistry have significant effect on the flame propagation. Considerable effort has been made within the community to correlate experimental data from different configurations (see [1–5], for example) in terms of parameters that are device-independent. However, to date those type of correlations have remained elusive and data appears to be sensitive to the flow configuration. Simulation has the potential to overcome the limitations of theory and experiment and provide new insights into the behavior of premixed flames. Our objective is to perform " first principles " simulations that incorporate detailed descriptions of chemistry and transport without the use explicit models for turbulence or turbulence/chemistry interaction. As examples here, we consider two turbulent premixed methane flame experiments: a V-flame anchored on a thin rod spanning a circular nozzle, and a piloted Bunsen flame anchored on a rectangular nozzle. These flames span a broad range and spatial and temporal scales. Simulation domains must be comparable to the window of the experimental diagnostics (order O(10) cm on side), while resolving the structure of the flame front that is less than 1 mm wide. Resolution requirements for the turbulent flow are less strigent but vary throughout the domain. Acoustic waves propagate at 10 5 cm/s in the hot products while typical fluid velocities are O(10 3) cm/s. With no turbulence present the flames overtake fuel at a laminar flame speed of 15-40 cm/sec. We have developed numerical methodolgy that combines a low-Mach number formulation combined with adaptive mesh refinement to exploit the variation of scales associated with premixed flame simulations. This combined approach, when implemented on parallel computing hardware, improves computational efficiency by several orders of magnitude. The reader is referred to [6] for details of the low Mach number model and its numerical implementation. …