Search Results

This paper describes an investigation of a method of implementing VVT with the use of a secondary valve in series with the conventional intake valve of the engine. The secondary valve is not required to withstand the temperatures and pressures of combustion, and therefore can be of relatively lightweight design, so that it is easier to adjust the timing of the secondary valve than that of the main valve. Experiments with such a valve installed in a production engine indicate that benefits of variable valve timing such as overlap optimisation and throttleless load control (4% Fuel benefits at 980 rpm and 1.5 bar IMEP) are attainable with this system.

A key point in three-way catalytic converter modeling problems is the definition of a possible chemical scheme able to represent the catalyzed process inside the converter, especially during transients. The lack of precise kinetic measurements during the transient thermal phase makes hard the choice of the kinetic expressions and, overall, of the chemical parameter values. To solve this problem here we propose the use of neural networks (NN) to model the reaction kinetics since a NN structure can provide enough degrees of freedom to capture all the significant features of the real system. Since the NN is embedded into the overall TWC dynamics, it cannot be trained through one of the standard method and some difficulties arise when dealing with the parameter tuning of this model, that are circumvented using a genetic algorithm (GA).

Electric vehicles today clearly represent the only solution fulfilling the zero emission vehicles (ZEV) standard. However, they are still not an equivalent alternative to gasoline driven cars due to the well known problems of today's batteries. The concept of a parallel hybrid drive line can be an optimal combination of both principles of propulsion in that the gasoline engine guarantees a wide range operation, while electric propulsion can be used within the restricted zero emission zones [20]. The parallel hybrid drive train described here has been realized at the Swiss Federal Institute of Technology (ETH), Zurich, Switzerland. For the first time a drive line consisting of a spark ignition engine, an electric asynchronous machine, a flywheel, and a wide range continuously variable transmission (CVT) is realized. An overall control system has been developed for the drive train. This drive line has been integrated in a multi purpose van for real road testing.

Emissions and performance parameters of a medium size, medium speed D.I. diesel engine with increased charge air pressure and reduced but fixed inlet valve opening period have been measured and compared to the standard engine. While power output and fuel consumption are slightly improved, nitric oxide emissions can be reduced by up to 20%. The measurements confirm the results of simulations for both performance and emissions, for which a quasidimensional model including detailed chemistry for nitric oxide prediction has been developed.

In the recent years, considerable progress has been achieved in the development of urea-SCR for nonstationary, mobile applications. Main challenges are the reduction of the catalyst volumes and the optimization of the dosing strategy to minimize transient ammonia emissions during load changes. Catalysts with increased cell density and enhanced intrinsic activity are one way to obtain smaller catalyst volumes. Another approach is the introduction of additional catalysts upstream and downstream of the SCR catalyst itself. Most effective is an oxidizing precatalyst that converts a part of NO to NO2, thus allowing a faster SCR reaction to occur. However, such a system requires the use of fuel with reduced sulfur content. Work performed at the Paul Scherrer Institute comprises of fundamental experiments in a laboratory scale reactor with synthetic exhaust gas and applied experiments on a diesel test stand.

Numerical simulations of in-cylinder soot evolution in the optically accessible heavy-duty diesel engine of Sandia National Laboratories have been performed with the multidimensional conditional moment closure (CMC) model using a reduced n-heptane chemical mechanism coupled with a two-equation soot model. Simulation results are compared to the high-fidelity experimental data by means of pressure traces, apparent heat release rate (AHRR) and time-resolved in-cylinder soot mass derived from optical soot luminosity and multiple wavelength pyrometry in conjunction with high speed soot cloud imaging. In addition, spatial distributions of soot relevant quantities are given for several operating conditions.

Particle emissions from four spark ignition engines were measured during steady state and transient chassis dynamometer tests. Transient Unified Drive Cycle tests were conducted at 0 °C and room temperature. Particle number, size distribution, active surface area, and photoelectric activity were determined. The results generally show low emission values for steady state operation of the warm engine. High particle concentrations are emitted in the first phase of the high-speed steady state testing. Once the engine is warmed up high emissions mainly occur during transient operation phases. The formation of nucleation mode particles is favored by the low concentration of carbonaceous soot, which offers volatile material little surface area for condensation.

Dual-fuel natural gas engines are seen as an attractive solution for simultaneous reduction of pollutant and CO2 emissions while maintaining high engine thermal efficiency. However, engines of this type exhibit a tradeoff between misfire as well as high UHC emissions for small pilot injection amounts and higher emissions of soot and NOX for operation strategies with higher pilot fuel proportion. The aim of this study was to investigate POMDME as an alternative pilot fuel having the potential to mitigate the emissions tradeoff, enabling smokeless combustion due to high degree of oxygenation, and being less prone to misfire due to its higher cetane number. Furthermore, POMDME can be synthetized carbon neutrally. First, characteristics of POMDME ignition in methane/air mixture and the transition into premixed flame propagation were investigated optically in a rapid compression-expansion machine (RCEM) by employing Schlieren and OH* chemiluminescence imaging.

The sooting propensity of dual-fuel combustion with n-dodecane pilot injection in a lean-premixed methane-air charge has been investigated using an optically accessible Rapid Compression-Expansion Machine (RCEM) to achieve engine-relevant pressure and temperature conditions at the start of pilot injection. A Diesel injector with a 100 μm single-hole coaxial nozzle, mounted at the cylinder periphery, has been employed to admit the pilot fuel. The aim of this study was to enhance the fundamental understanding of soot formation and oxidation processes of n-dodecane in the presence of methane in the air charge by parametric variation of methane equivalence ratio, charge temperature, and pilot fuel injection duration. The influence of methane on ignition delay and flame extent of the pilot fuel jet has been determined by simultaneous excited-state hydroxyl radical (OH*) chemiluminescence and Schlieren imaging.

This paper presents numerical simulations of in-cylinder soot evolution in the optically accessible heavy-duty diesel engine of Sandia Laboratories performed with the conditional moment closure (CMC) model employing a reduced n-heptane chemical mechanism coupled with a two-equation soot model. The influence of exhaust gas recirculation (EGR) on in-cylinder processes is studied considering different ambient oxygen volume fractions (8 - 21 percent), while maintaining intake pressure and temperature as well as the injection configuration unchanged. This corresponds to EGR rates between 0 and 65 percent. Simulation results are first compared with experimental data by means of apparent heat release rate (AHRR) and temporally resolved in-cylinder soot mass, where a quantitative comparison is presented. The model was found to fairly well reproduce ignition delays as well as AHRR traces along the EGR variation with a slight underestimation of the diffusion burn portion.

It is well-known that the spatial distribution of turbulence intensity and fuel concentration at spark-time play a pivotal role on the flame development within the pre-chamber in gas engines equipped with a scavenged pre-chamber. The combustion within the pre-chamber is in turn a determining factor in terms of combustion behaviour in the main chamber, and accordingly it influences the engine efficiency as well as pollutant emissions such as NOx and unburned hydrocarbons. This paper presents a numerical analysis of fuel concentration and turbulence distribution at spark time for an automotive-sized scavenged pre-chamber mounted at the head of a rapid compression-expansion machine (RCEM). Two different pre-chamber orifice orientations are considered: straight and tilted nozzles. The latter introduce a swirling flow within the pre-chamber. Simulations have been carried out using with two different turbulence models: Reynolds-Averaged Navier-Stokes (RANS) and Large-Eddy Simulation (LES).

In this work, numerical simulations of an automotive-sized scavenged pre-chamber mounted in an optically-accessible rapid compression-expansion machine (RCEM) have been carried out using two different turbulence models: Reynolds-Averaged Navier-Stokes (RANS) and Large-Eddy Simulation (LES). The RANS approach is combined with the G-equation combustion model, whereas the LES approach is coupled with the flamelet generated manifold (FGM) model for partially-premixed combustion. Simulation results are compared with experimental data in terms of OH* chemiluminescence in the main chamber. Both RANS and LES results were found to qualitatively reproduce the main features observed experimentally in terms of spatial flame development. Simulation results are further analysed by means of early flame propagation within the pre-chamber (related to the fuel and turbulence intensity distributions) and the ignition process in the main chamber.

Formation of soot in an auto-igniting n-dodecane spray under diesel engine relevant conditions has been investigated numerically. As opposed to research addressing turbulence-chemistry interaction (TCI) by coupling diffusive turbulence models with more sophisticated models in the context of Reynolds-Averaged Navier-Stokes equations (RANS), this study employs the advanced sub-grid scale k-equation model in the framework of a Large Eddy Simulation (LES) together with the uninvolved Direct Integration (DI) approach. A reduced n-heptane chemical mechanism has been employed and artificially accelerated in order to predict the ignition for n-dodecane accurately. Soot processes have been modelled with an extended version of the semi-empirical, two-equation model of Leung, which considers C2H2 as the soot precursor and accounts for particle inception, surface growth by C2H2 addition, oxidation by O2, oxidation by OH and particle coagulation.

Neutron imaging (NI) is an alternative non-destructive inspection technique compared to the well-known X-ray method. Although neutron imaging data look at a first glance similar to X-ray images it must be underlined that the interaction mechanism of the sample material with neutrons differs fundamentally. X-ray interaction with matter occurs with the electrons in the atomic shells whereas neutrons interact only with the atomic nuclei. Hence, both methods have a different and to great extent complementary contrast origin. Neutron imaging allows for a higher penetration through heavier elements (e.g. metals) whereas a high contrast is given for light elements (e.g. hydrogen). By the use of neutrons instead of X-rays exhaust after-treatment systems can be successfully examined non-destructively for their soot, ash, urea and coating distributions.

A computational study of the near-wall premixed flame propagation in homogeneous charge spark ignited engines is presented on the basis of a spectral concept accounting for flow-chemistry interaction in the flamelet regime. Flame surface enhancement due to wrinkling and modification of the local laminar flame speed due to flame stretch are the main phenomena described by the model. A high pass filter in the turbulent kinetic energy spectrum associated with the distance between the ensemble-averaged flame front location and the solid surface has been also introduced. In addition a probability density function of instantaneous flamelet positions around the above mean flame front location allows to consider statistical effects in a simplified way. Issues of temperature distribution within the boundary layer and associated heat losses, except for the concept of a thermal quenching distance, are thereby not explicitly taken into account.

The limiting values for NOx and HC concentrations in the exhaust gas of SI engines will be further lowered by legislation in many countries during the next years. This necessitates an improvement of the pollution control systems, which is achieved by including the dynamics of the three way catalyst into the control system. Before a control system can be designed, the dynamic behaviour of the exhaust after treatment system including the sensors has to be properly analyzed. As a first step a dynamic model of a solid-electrolyte oxygen sensor has been derived. It was the goal to obtain a better understanding of the cross sensitivities towards both reducing and oxidizing exhaust gas components such as H2, CO, O2 and NO. The model consists of three parts. Firstly, the porous protection layer, where only diffusion is assumed to occur, secondly the porous catalytic electrodes where the redox reactions take place and thirdly the solid electrolyte, where the electric potential is generated.

Compressed Natural Gas (CNG) engines have become a promising alternative to classical IC engines because of low pollutant and carbon dioxide emissions. This paper will first briefly summarize these advantages and then concentrate on the modeling and the control of CNG engines. In the modeling part, it will be shown which effects are similar to those observed in gasoline SI engines and what new sub-models are necessary. In the control part, the problem of sudden A/F ratio changes (for instance during the regeneration of NOx trap catalysts) will be considered. In order to avoid excessive NOx engine-out emission in these transients it is important to switch from lean to rich conditions within very few combustion cycles while keeping the engine torque constant (for comfort reasons). The paper presents a model of the most important phenomena associated with those transients and a feedforward control that meets the mentioned requirements.

Large eddy simulations (LES) of a port-injected 4-valve spark ignited (SI) engine have been carried out with the emphasis on the combustion process. The considered operating point is close to full load at 3,500 RPM and exhibits considerable cyclic variation in terms of the in-cylinder pressure traces, which can be related to fluctuations in the combustion process. In order to characterize these fluctuations, a statistically relevant number of subsequent cycles, namely up to 40, have been computed in the multi-cycle analysis. In contrast to other LES studies of SI engines, here the G-equation (a level set approach) has been adopted to model the premixed combustion in the framework of the STAR-CD/es-ICE flow field solver. Tuning parameters are identified and their impact on the result is addressed.

Knock is studied in a single cylinder direct injection spark ignition engine with variable intake temperatures at wide open throttle and stoichiometric premixed ethanol-air mixtures. At different speeds and intake temperatures spark angle sweeps have been performed at non-knocking conditions and varying knock intensities. Heat release rates and two zone temperatures are computed for both the mean and single cycle data. The in-cylinder pressure traces are analyzed during knocking combustion and have led to a definition of knocking conditions both for every single cycle as well as the mean engine cycle of a single operating point. The timing for the onset of knock as a function of degree crank angle and the mass fraction burned is determined using the “knocking” heat release and the pressure oscillations typical for knocking combustion.

The premixed flame growth period of about 1% of the cylinder mass burned has been theoretically investigated under typical homogeneous charge engine conditions. For this purpose various flame kernel development models have been tested against measured values of flame radius vs. time after ignition in a research engine. The flame kernel growth has been computed on the basis of a zero-dimensional model incorporating spark-induced energy, heat loss to the electrodes and flame curvature effects. Subsequently the transition phase from laminar to fully turbulent flame propagation is shown to depend strongly on the relationship between the turbulent kinetic energy spectrum and characteristic scales of the flame. We thereby make use of recently reported results of fundamental experiments on vortex-flamelet interaction, that yield typical vortex sizes for flame wrinkling and quenching.