Parameter sensitivity analysis via direct numerical simulation of boundary layer flow with heat release model as an analogy to bushfires.

Abstract: A thermal turbulent boundary layer (TTBL) flow was simulated via direct numerical simulation (DNS) using a temperature dependent heat release model to serve as an analogy to a bushfire. A number of characteristics of the heat release model such as the maximum heat release, and the corresponding temperature, are dependent on the adiabatic flame temperature and biomass activation energy, which in the model are formulated as non-dimensional parameters. The parameters within the heat release model were chosen to cover the range of adiabatic flame temperatures and activation energies respectively, as observed in real world bushfires, forming the basis of the parameter sweep. Multiple DNSs of the scenario were conducted to investigate the sensitivity of the flow evolution to the parameters within the heat release model via the analysis of flow features. Statistical analysis is performed to extract the biomass consumption rates, fire front velocities, and mean profiles to determine their sensitivities to the heat release model parameters. A mean pressure minimum located at the wall and fire front was identified as a key feature of the flow that is highly dependent on the heat release model parameters, and used to quantify the sensitivity of the flow to the heat release model. Keywords: Navier-Stokes equations, boundary-layer flows, direct numerical simulations, bushfires