Water injection in highly boosted gasoline direct injection (GDI) engines has become an attractive area over the last few years as a way of increasing efficiency, enhancing performance and reducing emissions. The technology and its effects are not new, but current gasoline engine trends for passenger vehicles have several motivations for adopting this technology today. Water injection enables higher compression ratios, optimal spark timing and elimination of fuel enrichment at high load, and possibly replacement of EGR. Physically, water reduces charge temperature by evaporation, dilutes combustion, and varies the specific heat ratio of the working fluid, with complex effects.Several of these mutually intertwined aspects are investigated in this paper through computational fluid dynamics (CFD) simulations, focusing on a turbo-charged GDI engine with port water injection (PWI). Different strategies for water injection timing, pressure and spray targeting are investigated. Two combustion modeling approaches are used and compared, the perfectly stirred reactor model with reduced chemical kinetics for a TPRF surrogate, and the G-equation turbulent combustion model. Combustion rate results are validated against available experimental engine data, in regular and knocking conditions. Multi-cycle simulations are required because of the wall film dynamics, and therefore have been performed to assess the effect of water injection strategies. The results of the study are an assessment of the optimal injection parameters, in terms of injector location, injection timing and primary atomization quality, for achieving the maximum effectiveness of water injection. Knock occurrences are in very good agreement with the experimental data and its suppression is demonstrated through the injection of water at a ratio of 30% to the fuel mass.

Among plasma-assisted ignition technologies, the Radio-Frequency (RF) corona family represents an interesting solution for the ability to extend the engine operating range. These systems generate transient, non-thermal plasma, which is able to enhance the combustion onset by means of thermal, kinetic and transport effects. Streamer-type RF corona discharge, at about 1 MHz, ignites the air-fuel mixture in multiple filaments, resulting in many different flame kernels. The main issue of this system is that at high electrode voltage and low combustion chamber pressure a transition between streamer and arc easily occurs: in this case transient plasma benefits are lost. A barrier discharge igniter (BDI), supplied with the same RF energy input, instead, is more breakdown-resistant, so that voltage can be raised to higher levels. In this work, a streamer-type RF corona igniter and a BDI were tested in a single-cylinder optical engine fueled with gasoline. Combustion behavior was characterized at stable, near-stable and unstable conditions (depending on IMEP coefficient of variation), increasing the air-fuel ratio starting from a near-stoichiometric mixture. For each test point the maximum allowable electrode voltage for both the igniters was used: in the streamer-type corona case it had to be limited because of the arc transition. Both indicating and imaging analysis were carried out, the latter by means of a high-speed camera that records the natural luminosity of early flames. Raw exhaust gas analysis was carried out as well. Results in terms of cycle-to-cycle variability, combustion duration, pollutant emissions, flame radii evolution and flame growth speed, as well as flame probability presence, are presented. The outcomes of this research can improve the knowledge on streamer-type corona and BDI and can be used to assess the operating range for the control parameters of these igniters.

It has been proved that Radio Frequency Corona, among other innovative ignition systems, is able to stabilize combustion and to extend the engine operating range in lean conditions, with respect to conventional spark igniters. This paper reports on a sensitivity analysis on the combustion behavior for different values of Corona electric control parameters (supply voltage and discharge duration). Combustion analysis has been carried out on a single cylinder PFI gasoline-fueled optical engine, by means of both indicating measurements and imaging. A high-speed camera has been used to record the natural luminosity of premixed flames and the obtained images have been synchronized with corresponding indicating acquisition data. Imaging tools allowed to observe and measure the early flame development, providing information which are not obtainable by a pressure-based indicating system. At different air-fuel ratio values, ranging from stoichiometric to lean limit, combustion stability has been evaluated for each combination of supply voltage and corona duration. Varying these parameters, it has been possible to assess the operating range of the corona igniter. In some limiting instances (far from normal operating range) not all the Corona igniter electrode tips were able to generate a discharge, worsening combustion onset: a statistical analysis of this phenomenon has been carried out. Cycle-resolved equivalent flame radii development and flame probability at fixed equivalent radius have been investigated for each condition. Finally, a suitable dataset of supply voltage and corona duration values, which guarantee a fast combustion onset and stable conditions, has been determined.