Computational methodology for non-evaporating spray in quiescent chamber using Large Eddy Simulation. (May 2018)

• Record Type: Journal Article
• Authors: Khan, Muhammad Mahabat; Hélie, Jerome; Gorokhovski, Mikhael
• Is Part Of: International journal of multiphase flow. Volume 102(2018)
• Page Start: 102
• Page End: 118
• Publication Date: 2018-05
• Subjects: Large Eddy Simulation -- Spray -- Euler–Lagrange -- Turbulent structures; Multiphase flow -- Periodicals; Écoulement polyphasique -- Périodiques; Multiphase flow; Periodicals; 620.1064
• DOI: 10.1016/j.ijmultiphaseflow.2018.01.025
• Languages: English
• ISSNs: 0301-9322

Abstract

Highlights: A comprehensive comparison of different LES SGS models in context of GDI non-evaporating sprays is performed. A new simple injection model along with two-way coupling is proposed. Complete LES spray simulation methodology is established. The numerical results are validated against experimental data. Abstract: Large Eddy Simulation of a single non-evaporating spray in a quiescent chamber is carried out in this paper. The spray is injected with a given angle, velocity and drop size in a quiescent chamber with ambient conditions. Initially, a detailed sensibility analysis of various injection velocity profiles and subgrid carrier phase models, in the absence of droplet breakup and collision models, is carried out to capture the detailed turbulent flow features. The resulting spray behaviour is compared to mono-phase jet characteristics. The results provide basis for the selection of suitable turbulence model and injection profiles of the droplets for performing spray simulation in quiescent environmental conditions. Lastly the spray simulations are performed under realistic engine like conditions. For this purpose appropriate turbulence and droplet injection models along with droplet breakup model and collision model are used. The results are compared with the experimental data. The droplet size and fluctuating velocities are found to match reasonably well with experimental data.