Experimental characterization of the injected mass variation in a high-pressure GDI injector operating with a multiple injection strategy

Abstract

Increasingly stringent limits to pollutants released by Internal Combustion Engines pushed the automotive research to develop technologies to reduce fuel consumption and emissions. Higher injection pressures are beneficial to accelerate the atomization phase, reducing the particulate matter and unburned hydrocarbon emissions. However, the spray protrusion inside the combustion chamber is enhanced and, consequently, the generation of a thick wall film, which tends to increase the latter emissions. Thus, multiple-injection strategies might be beneficial for both the atomization rate and the spray penetration, owing to a stratified charge inside the chamber. This paper investigates the effect of the adoption of multiple-injection strategies on the behaviour of a GDI injector operating in high injection pressure conditions. The resulting injected mass is influenced by electrical phenomena on the excitation circuit, which mainly depend on the relative time between the end of the first injection and the start of the following. Hence, the total amount of fuel injected with the multiple-injection pattern will differ from its nominal value. In this work, a specific experimental layout was developed to characterize the behaviour of the injector in different operating conditions and quantify the deviation between actual and nominal injected mass. The impact of the magnetized coils on the overall injected mass has been captured referring to the modification of the shape of the driving current profile with respect to the nominal one. Then, a correlation which considers the electric charge variation on the coils has been implemented to model the phenomenon and, consequently, to counterbalance the electro-magnetic effect on the injected mass. The resulting strategy successfully allowed to reduce the difference between the actual and target fuel mass from up to 30% to almost 5%, owing to its implementation on the injection control system to automatically correct the injection commands and compensate the fuel mass deviations.