A novel spray model validation methodology using liquid-phase extinction measurements

Physical spray models employed in engine computational fluid dynamics (CFD) simulations are not yet fully predictive; therefore, the breadth of conditions under which these simulations yield valid predictions depends strongly on the "tuning" of these models against available spray measurements. Often, these models are validated and calibrated against spray images based on the elastic scattering of light, or Mie scattering, from liquid structures and droplet clouds. However, these measurements do not typically detect the absolute liquid boundary, so employed computational metrics used to define the liquid boundary in the modeled spray can be physically inconsistent with that detected in Mie-scatter images. To more robustly validate fuel spray model predictions against light scattering measurements, direct comparisons can be made between predicted and measured light scattering intensity signals. Such a comparison provides a more quantitative validation of the liquid phase fuel boundary and further offers the potential to validate local spray structure. In this work, we apply the Lorentz-Mie solution to Maxwell's equations to predict extinction signals due to elastic light scattering, informed by droplet diameter and number density distributions, within a predicted diesel spray. The predicted extinction is compared to experimental results from diffused back-illumination and single line-of-sight extinction measurements to generate a calibrated model of the Engine Combustion Network "Spray A" condition that replicates the measured centerline extinction profile. This spray model is used to inform liquid volume fraction thresholds to similarly define the detected liquid boundary from Mie-scatter images.