Aymeric Lamorlette

Analytical Modeling of Solid Material Ignition Under a Radiant Heat Flux Coming From a Spreading Fire Front

This study aims at characterizing ignition of solid targets exposed to spreading fire fronts. In order to model radiant heat fluxes on targets in a realistic way, polynomial heat fluxes are chosen. Analytical solutions for the solid surface temperature evolution regarding different time-varying heat fluxes are discussed for high thermal inertia solids using a mathematical formalism, which allows for the methodology to be extended to the case of low thermal inertia. This formulation also allows calculation of ignition times for more realistic time-dependent fluxes than previous studies on the topic, providing a more general solution to the problem of solid material ignition. Polynomial coefficients are then obtained fitting heat flux coming from absorbing-emitting flames. A characterization of solid material ignition times regarding fire front rate of spread (ROS) is finally performed, showing the need to accurately model heat flux variations in ignition time calculations.

An improved non-equilibrium model for the ignition of living fuel

This paper deals with the modelling of living fuel ignition, suggesting that an accurate description using a multiphase formulation requires consideration of a thermal disequilibrium within the vegetation particle, between the solid (wood) and the liquid (sap). A simple model at particle scale is studied to evaluate the flux distribution between phases in order to split the net flux on the particles into the two sub-phases. An analytical solution for the split function is obtained from this model and is implemented in ForestFireFOAM, a computational fluid dynamics (CFD) solver dedicated to vegetation fire simulations, based on FireFOAM. Using this multiphase formulation, simulations are run and compared with existing data on living fuel flammability. The following aspects were considered: fuel surface temperature, ignition, flaming combustion time, mean and peak heat release rate (HRR). Acceptable results were obtained, suggesting that the thermal equilibrium might not be an acceptable assumption to properly model ignition of living fuel.

A dimensional analysis of forest fuel layer ignition model: Application to the ignition of pine needle litters

This article deals with the physical modelling of forest fuel layer ignition. A model based on momentum-, fluid- and solid-phase energy equations is written for a fuel layer and a dimensional analysis is performed. This analysis allows to enlighten on two relevant dimensionless groups regarding the dimensionless time to ignition of a fuel layer and also provides a suited scaling for the fluid velocity inside the fuel layer during ignition. A correlation for the time to ignition is then fitted on experimental data obtained using an FM-Global fire propagation apparatus for different pine species with a closed basket. A good agreement is found, emphasizing the relevance of the dimensionless groups and the thermally thick behaviour of the solid particles during the ignition process under incident radiant heat flux as low as 8 − 12 kW m − 2 .

Quantification of convective heat transfer inside tree structures

Convective heat transfer between a vegetal structure and its surrounding medium remains poorly described. However, for some applications, such as forest fire propagation studies, convective heat transfer is one of the main factors responsible for vertical fire transitions, from ground level to the tree crowns. These fires are the most dangerous because their rates of spread can reach high speeds, around one meter per second. An accurate characterization of this transfer is therefore important for fire propagation modelling. This study presents an attempt to formulate a theoretical modelling of the convective heat transfer coefficient for vegetal structures generated using an Iterated Function Systems (IFS). This model depends on the IFS parameters. The results obtained using this approach were compared with previously computed numerical results in order to evaluate their accuracy. The maximal discrepancies were found to be around 12% which proves the efficiency of the present model.