ID 317

Turbulent statistical transition from Euler to Lagrange using droplet velocity PDF

 

Abstract:

The Euler-Lagrange Spray Atomization model, namely ELSA [1, 2], is a multi-scale approach suitable to perform Large Eddy Simulations (LES) together with the possibility to recover Direct Numerical Simulation (DNS) features for well resolved interfaces. Recent validations with experimental and DNS data were made within the primary break-up [3, 4, 5]. Nevertheless, the link between secondary atomization (where liquid sheets break into ligaments and bag-like structure) and dilute or dispersed zone (where spray of droplets are formed) is still an open question. One of the major challenge within the transition from dense zone (Euler) to dilute zone (Lagrange) is the droplet size and velocity probability density functions (PDF) in turbulent jets. Normally, in diffuse interface models, the averaged mixture velocity and surface interface are employed to set up the lagrangian droplet. Nonetheless, such approach is far for being realistic. Consequently, among novel strategies e.g. 1) local turbulent statistic to improve averaged velocity mixture, 2) Quasi-Multiphase Euler approach [6] to recover averaged velocity in the liquid phase, and 3) local droplet PDF. Theses techniques would allow a local statistical transition of information from Euler to Lagrange approach. Indeed, an extraction from DNS data has been made, that introduces a new formulation of the drop size distribution (DSD) based on the liquid-gas surface curvature rather than the spherical droplet diameter [7]. The aim of this work is to enhance the coupling Euler-Lagrange in the atomization model, namely ELSA, by including the DSD from DNS approach [7] and velocity PDF from turbulent statistic.