The mixture composition heavily influences the combustion process of Port Fuel Injection (PFI) engines. The local mixture air-index at the spark plug is closely related to combustion instabilities and the cycle-by-cycle Indicated Mean Effective Pressure (IMEP) Coefficient of Variation (CoV) well correlates with the variability of the flame kernel development. The needs of reducing the engine emissions and consumption push the engine manufactures to implement techniques providing a better control of the mixture quality in terms of homogeneity and variability.
Simulating the mixture formation of a PFI engine by means of CFD techniques is a critical issue, since involved phenomena are highly heterogeneous and a two phase flow must be considered. The aim of the paper is to present a multi-cycle methodology for the simulation of the injection and the mixture formation processes of high performance PFI engine, based on the validation of all the main physical sub-models involved.
A semi-empirical methodology for the correct initialization of the Lagrangian spray is presented in the paper. The spray-wall interaction sub-models are usually based on semi-empirical correlation. In this paper the Kuhnke model was tuned by means of experimental data, chosen coherently with the spray phenomena taking place in the considered engine. Since the liquid wall film plays a key role in the mixture formation of PFI engines, an accurate representation of the wall film dynamics was enforced by the solution of the liquid film momentum equation. The gas flow dynamics in the intake port strongly interact with the liquid fuel evolution and droplet breakup, thus in this work a multi-cycle methodology for the evaluation of the mixture inside the cylinder was proposed. In order to validate the simulation results, an optical access has been created on the engine airbox, allowing to use a fast camera to capture images of the actual injection process. The comparison of simulated and acquired images confirmed that both the gas and liquid fuel dynamics have been correctly reproduced.
The evaluation of the injection timing influence on the engine performance was finally accomplished.