A combination of computational fluid dynamics models implemented in a RANS flow solver are used to investigate the characteristics of gasoline sprays injected from three different types of high-pressure nozzles for direct injection spark-ignition engines: the swirl-pressure, the multi-hole and the outwards opening pintle injectors. The initial conditions required for the spray development are determined by solving for the flow distribution inside the nozzles. The subsequent spray development is predicted using a Eulerian-Lagrangian stochastic methodology adopted for droplet motion calculations in numerical grids with cell size comparable to that of the droplets. Many of the fundamental physical processes taking place during the spray development are incorporated into the model, including link with the internal nozzle flow, liquid-core and liquid-sheet atomisation, droplet aerodynamic break-up, turbulent dispersion, vaporisation, drop-to-drop interactions and wall impingement. Model validation takes place against experimental results available for fuel injection into ambient air, a constant volume chamber operating at elevated pressures and temperatures and the cylinder of a transparent multi-valve direct injection engine. Results demonstrate that simulation of the internal nozzle flow and its link with the spray is critical for accurate prediction of the characteristics of the developing sprays as function of the design of the fuel injection system used
Modelling of sprays from high-pressure nozzles for direct injection gasoline engines
TONINI, Simona;COSSALI, Gianpietro;MARENGO, Marco
2006-01-01
Abstract
A combination of computational fluid dynamics models implemented in a RANS flow solver are used to investigate the characteristics of gasoline sprays injected from three different types of high-pressure nozzles for direct injection spark-ignition engines: the swirl-pressure, the multi-hole and the outwards opening pintle injectors. The initial conditions required for the spray development are determined by solving for the flow distribution inside the nozzles. The subsequent spray development is predicted using a Eulerian-Lagrangian stochastic methodology adopted for droplet motion calculations in numerical grids with cell size comparable to that of the droplets. Many of the fundamental physical processes taking place during the spray development are incorporated into the model, including link with the internal nozzle flow, liquid-core and liquid-sheet atomisation, droplet aerodynamic break-up, turbulent dispersion, vaporisation, drop-to-drop interactions and wall impingement. Model validation takes place against experimental results available for fuel injection into ambient air, a constant volume chamber operating at elevated pressures and temperatures and the cylinder of a transparent multi-valve direct injection engine. Results demonstrate that simulation of the internal nozzle flow and its link with the spray is critical for accurate prediction of the characteristics of the developing sprays as function of the design of the fuel injection system usedPubblicazioni consigliate
Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo