Understanding the development of a reacting fuel jet inside an automotive-size diesel engine using optical and laser-based diagnostics

摘要

The fuel penetration and reacting diesel jet development have been studied in a small-bore optical engine to improve the understanding of a swirl-influenced, wall-interacting diesel flame. The optical access to the engine combustion chamber was made possible via multiple quartz windows positioned in a cylindrical piston bowl and cylinder liner. Using the common-rail fuel injection system of the engine, the fuel injection was executed for long duration, creating negative ignition dwell conditions in which the start of combustion occurs before the end of injection. A single-hole nozzle was used to isolate the jet-wall interaction from jet-jet interactions while limiting the in-cylinder pressure below the burst-pressure of quartz windows. Planar laser-induced fluorescence imaging of hydroxyl (OH-PLIF), fuel-PLIF, and line-of-sight integrated chemiluminescence imaging were performed for various combustion stages identified by the in-cylinder pressure traces and apparent heat release rates. These include stages of vaporising fuel penetration, low-temperature reaction, and high-temperature reaction. The fuel-PLIF images show that the fuel penetration was strongly influenced by a swirl flow with the wall-jet penetration on the up-swirl side being shorter than that of the down-swirl jet. During the low-temperature reaction, cool flame chemiluminescence appears in the wall-jet head region. Interestingly, this region is where a turbulent ring-vortex is formed due to jet-wall interactions, suggesting that locally enhanced mixing induced the first-stage ignition. The OH-PLIF images show that the second-stage, high-temperature reaction starts to occur and then expand drastically in the same wall-jet head region. Since the reaction occurs in the wall-jet region, the swirl flow impacts the high-temperature reaction significantly, as evidenced by more intense OH signals in the down-swirl jet. This is due to the influence of the swirl flow on the mixing process, leading to relatively richer mixtures on the down-swirl side. Upon the end of fuel injection, the heat release rate declines and the OH-PLIF signals slowly diminish.How the variation in injection pressure influences the combustion processes of a wall-interacting diesel jet has also been investigated. The cool-flame images together with the apparent heat release rate suggest that the low-temperature reaction still emerges from the wall-interacting jet head region but it becomes stronger with increasing injection pressure due to the better air-fuel mixing at the enhanced turbulent ring-vortex. The influence of in-cylinder swirl flow on the OH* chemiluminescence signals was again observed such that the high-temperature reaction in the down-swirl side of the jet occurs earlier than that in the up-swirl side of the jet regardless of the injection pressure. Moreover, the second-stage ignition on the down-swirl side of the jet is also found to be stronger than the up-swirl side of the jet initially. However, as the injection pressure increases and the high temperature reaction matures, the spread and magnitude of the up-swirl OH* chemiluminescence signals become comparable to the down-swirl signals owing to the increased injection momentum overcoming the swirl flow. The OH-PLIF signals indicate that the high-temperature reaction zone continues to grow in the turbulent ring-vortex region where the cool-flame signals were detected at earlier timings. The expansion of wall-jet head OH signals shows an interestingly growing trend with increasing injection pressure, which can be explained by a stronger ring-vortex due to the increased injection momentum.At selected operating conditions of 100 MPa common-rail pressure and long 2.04-ms injection duration, planar laser-induced incandescence (PLII) imaging has been performed to clarify soot processes within the wall-interacting jet. Once again, a single-hole nozzle was used to isolate the jet-wall interaction from jet-jet interactions and to apply long injection duration corresponding to high-load engine operating conditions in which soot formation is particularly problematic. Compared to the previous experiments, two major changes were made in fuel and piston design. As opposed to a conventional diesel fuel used in the previous experiments, the soot diagnostics were conduced using methyl decanoate, a surrogate fuel with low-sooting propensity, to reduce laser attenuation. In addition, the piston bowl design was modified to include a curved bowl wall to enhance the fuel jet penetration back towards the nozzle, which is closer to the conditions in most production engines. Laser-based images show that the fuel impinges on the bowl wall soon after the start of injection and then bounces off along the wall forming a wall-interacting jet. The fuel jet continues to travel along the bowl wall as well as the bottom surface of the piston bowl. Although the latter motion was not significant in the previous experiments, with the new curved bowl-wall, the fuel penetration back towards the nozzle was clearly observed. During the premixed burn phase of diesel combustion, the high-temperature reaction starts to occur at the leading edge of the penetrating jet back towards the nozzle, initially near the jet axis and then spreads in the radial direction. During the mixing-controlled burn phase, the high-temperature reaction zone fills up the entire combustion chamber and the soot formation starts to occur in the rich area near the wall impingement point. The soot then flows along the bowl wall in both up-swirl and down-swirl directions. Throughout this phase, these soot pockets are surrounded by OH, which disappear altogether at subsequent crank angle locations suggesting the soot oxidation by OH radicals. However, some soot pockets are transported into the centre of bowl due to the downward movement of the piston and persist for long as there are no active OH radicals. To conclude, these major findings made on the temporal and spatial evolution of a wall-interacting diesel jet, its variations with increasing injection pressure, and the soot concentration within the jet are summarised by illustrating regions of fuel, low- and high-temperature reaction, as well as soots for various crank angle locations during a firing cycle of the engine.