16th AIAA Computational Fluid Dynamics Conference · 2003 · 20 citations · 11 references
EngineeringChemical EngineeringHigh-performance ComputingCombustionComputational Fluid DynamicsComputational Heat TransferFire DynamicsFire ModelingFire SimulationHeat Transfer In FiresLarge-eddy SimulationTurbulent CombustionFiresHeat Transfer MechanismsPuffingPuffing PhenomenaSoot Formation KineticsSoot Formation ModelingTime ScalesVortical Flow
Large-scale hydrocarbon pool-fires (liquid hydrocarbon fires of pool diameters > 10m) are difficult to analyze experimentally because of the scale of the events. Massively parallel computing using thousands of processors offers a vehicle to deliver simulation-science based analysis of such large-scale events. On these scales the fires are turbulent; that is, dynamic vortical structures are present. The fire chemistry requires reaction mechanisms that allow for soot formation, growth and oxidation. Radiation, the dominant mode of heat transfer, is strongly affected by the presence of soot. Large Eddy Simulation (LES) provide temporally and spatially resolved information over time scales that include the well known puffing frequency and overall burning time for these fires and over spatial scales that include the vortical structures on the order of 10cm in scale and larger. Smaller spatial and temporal scale (subgridscale) processes are modeled. Verification and validation of the LES fire simulation tool with its subgridscale models provides some confidence that simulations replicate observables.
11