Accounting for soot particle morphology in flame thermal radiation and optical diagnostics in complex systems – ASTORIA

The methodology proposed in ASTORIA requires a strong expertise in high fidelity Direct Numerical Simulation (DNS) and Large Eddy Simulation (LES) of sooting flames in complex geometries, Diffusion Limited Cluster-Cluster Aggregation (DLCA) approach for aggregate generation and accurate radiation simulation of both the burnt gases released by combustion and the soot aggregates. Both Discrete Ordinate Method (DOM) and Monte Carlo Methods (MCM) will be employed, together with the Discret Dipole Approximation to evaluate the radiative properties of individual aggregates. To the best of our knowledge, such coupled simulations have never been performed and give to the ASTORIA project significant originality and innovation in comparison with other current projects in the field.

-

-

-

The ASTORIA project aims to develop innovative numerical tools for a significant improvement of the CFD modelling of both soot formation and radiative properties in complex combustion systems. Indeed, soot modelling is a difficult task involving complex chemical schemes, physical and radiative processes. Nevertheless, there is a strong need for a better control of the soot formation process, in order to optimize the radiative properties of combustion systems rand also to reduce the particle matter emissions. To gain a sufficient accuracy, models need to take into account the particle size and fractal aggregate morphology. For that purpose, ASTORIA aims to develop or improve numerical approaches both for combustion and radiation in complex geometries. First, ASTORIA will propose an alternative method based on a Lagrangian semi-deterministic method to existing Sectional Methods and Methods Of Moments allowing to compute the evolution of the PM population and properties, including a detailed gaseous chemistry. Second, realistic soot sizes and morphologies will be produced by developing an aggregation code relying on the surface growth and oxidation/fragmentation processes, taking into account physical local conditions along particle trajectories. Finally, radiative properties of the so-modelled realistic aggregates will be determined in the infrared and visible domains. The results will be implemented in highly sophisticated radiative transfer solvers in order to produce accurate evaluation of the radiative transfer in the combustion chamber caused both by burnt gases and soot particles. All these new model developments will be integrated in a combustion code and several radiation solvers with increasing accuracy to improve the prediction of emitted PM number and shape in industrial applications, which is the global objective of ASTORIA. Such developments will also allow to compute radiation in the visible frequency range, in order to reproduce numerically laser-based diagnostics such as static light scattering, line of sight extinction or laser induced incandescence. This will pave the way to a more direct comparison between simulations and experiments for a better understanding and prediction of PM emission. The improved knowledge and description of soot radiative properties will also increase the accuracy of laser-based diagnostics and climate models.
All these novel developments will be systematically validated based on existing documented flames (LII database of academic flames, ISF3-Target swirled turbulent sooting flame) and in the MICADO test rig studied within the H2020 SOPRANO project. An additional experimental campaign of planar laser light scattering will be conducted in MICADO to validate the numerical approach used to compute the digital image of the flame.
The ASTORIA project gathers four partners (CERFACS, CORIA, ONERA, RAPSODEE) expert in high-fideliy CFD of sooting reacting flows, soot aggregate generation and radiation modeling.