ISSN 打印: 1044-5110
ISSN 在线: 1936-2684

# 雾化与喷雾

DOI: 10.1615/AtomizSpr.v8.i3.50
pages 307-347

## LINKING NOZZLE FLOW WITH SPRAY CHARACTERISTICS IN A DIESEL FUEL INJECTION SYSTEM

C. Arcoumanis
School of Engineering, City University, London
Manolis Gavaises
School of Mathematics, Computer Science, and Engineering, City University London, Northampton Square, EC1V 0HB London, United Kingdom

### ABSTRACT

A computer model simulating the flow in fuel injection systems and the characteristics of diesel sprays injected from multihole orifice-type injectors has been developed and validated against experimental data. The injection conditions were modeled by solving the wave dynamics in the fuel injection equipment (FIE) using a one-dimensional model The flow in the sac volume was treated as a three-dimensional one for the noncavitating cases, in order to identify the basic flow distribution at the exit of the nozzle holes. For the cavitating cases, a one-dimensional sac volume flow model was developed; correlations giving the discharge coefficient of the injection holes were used in order to predict the quantity of fuel injected. Both inclined and vertical injectors can be modeled, since different flow rates are predicted for the different holes of multihole nozzles, depending on the position of the injection hole relative to the sac volume and the needle seat axis. The injection velocity calculation was based on the effective area (area at the hole exit occupied by liquid due to the presence of cavitating bubbles), which is calculated as a function of the flow conditions in the sac volume, the hole geometric characteristics, and the back pressure. Following the initiation of fuel injection, with droplet velocity, size, hole effective area, and level of turbulence at the nozzle exit calculated from the FIE model, an existing three-dimensional computational fluid dynamics (CFD) diesel spray model was extended and used for the prediction of the spray characteristics. This model is based on the Eulerian-Lagrangian stochastic particle technique; the gas phase is simulated by solving numerically the full Navier-Stokes equations, while the liquid phase is modeled using a Lagrangian particle tracking approximation. Spray submodels were used to represent the various physical phenomena taking place during the spray development. Emphasis was placed on the effect of the nozzle flow on the disintegration of the emerging liquid jet. Different jet atomization models were used in order to predict the spray characteristics at the closest point to the hole exit where experimental data were available. The aerodynamic-induced atomization, the jet turbulence-induced atomization, and a newly developed model for the cavitation-induced atomization were the three mechanisms considered responsible for the disintegration of the liquid jet. The latter mechanism was combined with a correlation representing the radial distribution of the droplets in the spray cone angle. Droplet secondary breakup, droplet collisions, and droplet turbulent dispersion were taken into account as the droplets penetrate into the surrounding gas. Quiescent atmospheric gas conditions have been selected for model validation in order to concentrate on the effect of the nozzle flow on the spray characteristics without the added complications induced by high-density effects or gas motion. The computational results were extensively compared with available experimental data, and confirm that hole cavitation enhances atomization. These results include a typical set of predictions for the flow in the FIE, the velocity, pressure, and turbulence kinetic energy distributions in the sac volume for noncavitating transient flow conditions, the fuel injected per injection hole, and the effective hole area for the one-dimensional sac volume flow model, and the droplet temporal and spatial characteristics (velocity and size) for the spray model Two different injection conditions were examined, corresponding to pump speeds of 600 and 1200 rpm. From the comparison between the two different sets of results, it can be concluded that the combined FIE-spray model is capable of predicting the spray characteristics without a priori knowledge of the flow characteristics in the sac volume and at the exit of the injection holes.

### Articles with similar content:

THREE-DIMENSIONAL SIMULATION OF EFFERVESCENT ATOMIZATION SPRAY
Atomization and Sprays, Vol.19, 2009, issue 1
Jian-Zhong Lin, Ze-Fei Zhu, Hong-Bing Xiong
EXPERIMENTAL AND NUMERICAL MODELING STUDY FOR IRRIGATION GUN WATER JET
Atomization and Sprays, Vol.18, 2008, issue 4
N. Kadem, R. Borghi, A. Tchiftchibachian
EXPERIMENTAL AND NUMERICAL ANALYSIS OF SPRAY DISPERSION AND EVAPORATION IN A COMBUSTION CHAMBER
Atomization and Sprays, Vol.19, 2009, issue 10
Andreas Dreizler, Johannes Janicka, Amsini Sadiki, M. Hage, Mouldi Chrigui
VELOCITY MEASUREMENTS OF DENSE DIESEL FUEL SPRAYS IN DENSE AIR
Atomization and Sprays, Vol.14, 2004, issue 6
F. Obermeier, C. Kirmse, H. Chaves
MODELING SPRAY ATOMIZATION WITH THE KELVIN-HELMHOLTZ/RAYLEIGH-TAYLOR HYBRID MODEL
Atomization and Sprays, Vol.9, 1999, issue 6
Rolf D. Reitz, Jennifer C. Beale