Search Results

The gradual global tightening of emission legislation for particulate matter emissions requires the development of new gasoline engine exhaust aftertreatment systems. For this reason, the development of gasoline direct injection engines aims at the reduction of particulate emissions by application of a Gasoline Particulate Filter (GPF). The regeneration temperature of GPF depend on soot reactivity towards oxidation and therefore on particle properties. In this study, the soot reactivity is correlated with nanostructural characteristics of primary gasoline particles as a function of specific engine injection parameters. The investigations on particle emissions were carried out on a turbocharged 4-cylinder GDI-engine that allows the variation of injection parameters. The emitted engine soot particles have been in-situ characterized towards their number and size distribution using an engine exhaust particle sizer (EEPS).

In this work, the potential of laser-induced ignition to improve combustion initiation and heat release in a direct-injection engine is investigated by a combined experimental and numerical investigation. Laser ignition is studied in fuel/air mixtures with homogeneous equivalence ratio fields. The results provide knowledge about minimum ignition energies and the ignition limits of laser-induced ignition. Furthermore, in mixtures with nominally identical conditions, statistical variations of the ignition success are observed experimentally. These variations can be explained, based on numerical simulations, by fluctuations in the strain rate in the turbulent in-cylinder flow. Additionally, laser ignition in engines with a spray-guided combustion mode, with strongly inhomogeneous fuel/air mixtures, was investigated.

Investigations of the fuel injection processes in a spark ignition direct injection engine have been performed for two different fuels. The goal of this research was to determine the differences between isooctane, which is often used as an alternative to gasoline for optical engine investigations, and a special, non-fluorescing, full boiling range multicomponent fuel. The apparent vaporization characteristics of isooctane and the multicomponent fuel were examined in homogeneous operating mode with direct injection during the intake stroke. To this end, simultaneous Mie scattering and planar laser induced fluorescence imaging experiments were performed in a transparent research engine. Both fuels were mixed with 3-Pentanone as a fluorescence tracer. A frequency-quadrupled Nd:YAG laser was used as both the fluorescent excitation source and the light scattering source.

The spray propagation and disintegration is investigated in a pressure chamber. With Particle Image Velocimetry the direction and velocity of both, fuel droplets and induced gas flow are detected. By means of shadow photographs the spray cone geometry is visualized. To verify the predictions made of the measurements mentioned above and to rate the quality of the tuning of the parameters in-cylinder gas flow, injection pressure, position of Injector and position of spark plug under real engine conditions, a fast gas sampling valve is used in three different engines. The in-cylinder gas temperature and the soot concentration are measured crank angle resolved by means of the Two-Colour-Method in a 1-cylinder GDI-engine. The soot concentration and temperature show the influence of the injection pressure on emissions like soot and nitric oxide.

High frequency ignition (HFI) and conventional transistor coil ignition (TCI) were investigated with an optically accessible single-cylinder research engine to gain fundamental understanding of the chemical reactions taking place prior to the onset of combustion. Instead of generating heat in the gap of a conventional spark plug, a high frequency / high voltage electric field is employed in HFI to form chemical radicals. It is generated using a resonant circuit and sharp metallic tips placed in the combustion chamber. The setup is optimized to cause a so-called corona discharge in which highly energized channels (streamers) are created while avoiding a spark discharge. At a certain energy the number of ionized hydrocarbon molecules becomes sufficient to initiate self-sustained combustion. HFI enables engine operation with highly diluted (by air or EGR) gasoline-air mixtures or at high boost levels due to the lower voltage required.

As an important means of engine development and optimization, modelbuilding plays an increasingly important role in reducing carbon dioxide emissions of the internal combustion engines (ICEs). However, due to the non-linearity and high dimension of the engine system, a large amount of data is required to obtain high model accuracy. Therefore, a modelling approach combining the experimental data and prior knowledge was proposed in this study. With this method, an artificial neural network (ANN) model simulating the engine brake specific fuel consumption (BSFC) was established. With mean square error (MSE) and Kullback-Leibler divergence (KLD) serving as the fitness functions, the 86 experimental samples and constructed physical models were used to optimize the ANN weights through genetic algorithms.