This content is not included in your SAE MOBILUS subscription, or you are not logged in.

LES of Diesel and Gasoline Sprays with Validation against X-Ray Radiography Data

Journal Article
2015-01-0931
ISSN: 1946-3952, e-ISSN: 1946-3960
Published April 14, 2015 by SAE International in United States
LES of Diesel and Gasoline Sprays with Validation against X-Ray Radiography Data
Sector:
Citation: Wang, Z., Swantek, A., Scarcelli, R., Duke, D. et al., "LES of Diesel and Gasoline Sprays with Validation against X-Ray Radiography Data," SAE Int. J. Fuels Lubr. 8(1):147-159, 2015, https://doi.org/10.4271/2015-01-0931.
Language: English

Abstract:

This paper focuses on detailed numerical simulations of direct injection diesel and gasoline sprays from production grade, multi-hole injectors. In a dual-fuel engine the direct injection of both the fuels can facilitate appropriate mixture preparation prior to ignition and combustion. Diesel and gasoline sprays were simulated using high-fidelity Large Eddy Simulations (LES) with the dynamic structure sub-grid scale model. Numerical predictions of liquid penetration, fuel density distribution as well as transverse integrated mass (TIM) at different axial locations versus time were compared against x-ray radiography data obtained from Argonne National Laboratory. A necessary, but often overlooked, criterion of grid-convergence is ensured by using Adaptive Mesh Refinement (AMR) for both diesel and gasoline. Nine different realizations were performed and the effects of random seeds on spray behavior were investigated. Additional parametric studies under different ambient and injection conditions were performed to study their influence on global and local flow structures for gasoline sprays. It is concluded that LES can generally well capture all experimental trends and comes close to matching the x-ray data. Discrepancies between experimental and simulation results can be correlated to uncertainties in boundary and initial conditions such as rate of injection and spray and turbulent dispersion sub-model constants.