Study of the role of engine control in the value chain for biofuels in modern "ultra clean" engines
MetadataVis full innførsel
The climate changes caused by the combustion of fossil fuels in the transportation sector have, along with a decrease in fossil fuel reserves, resulted in an increased interest in developing alternative fuels. Biofuels are one of the most prominent options and with an expansion in the use of these fuels, it is important to understand all aspects of the environmental effect they impose.First generation biofuels are the commercial available biofuels today. Since their feedstock origins from food and oil-seed crops there is a great skepticism around their sustainability, creating a focus on research and development of second generation and advanced biofuels. These fuels are more environmental beneficial, however expensive to produce due to a more complex structure of the feedstocks. This requires a more advanced conversion technology, and additionally research is needed to make these fuels economically competitive. Fischer Tropsch Diesel (FTD) is one of the most promising second generation biofuels, yielding great reduction in greenhouse gas (GHG) emissions and fossil fuel consumption when looking at its whole life cycle. There are uncertainties around the combustion characteristics and the end use emissions of the different biofuels. This especially applies to the use of biofuels in blend with diesel, which is the most common form of utilizing biofuels today. In order to increase the use of these fuels the uncertainties need to be fully explored to ensure their sustainability. This research has traditionally been performed experimentally, but later years computational simulation has arisen as a powerful tool. It saves time and cost, and can also reveal information not possible to obtain by experiments in real engines. In order to obtain reliable results from the simulations, both high quality physical and chemical models are required. In this thesis a Stochastic Reactor Model (SRM) is used, where the volume in the cylinder is divided into a number of smaller volumes. These are known as particles and have their own chemical composition, temperature and mass. Since the SRM is a 0-dimensional model no characteristics regarding the position in space are given, but a Probability Density Function (PDF) gives a distribution of the properties of the particles and enables them to mix and exchange heat with the cylinder walls. By doing so the model takes into account in-homogeneities and turbulence. The engine type used is a direct injection compression ignition (DICI) engine where fuel is directly injected, hence good models are required for both the mixing process as well as the chemical kinetics. The chemical models used for simulations should withhold the same characteristics as the original fuels, however due to time limitations for the computational calculations less important species and reaction paths should be eliminated. This is done through a reduction process, where there always exists a trade-off between the quality of the model, and the time consumption. The chemical models applied in this thesis are substitutions and simplifications of the original fuels, namely diesel and FTD. N-heptane (nC7H16) has been used as a surrogate fuel for diesel, while the fuel composition representing FTD is 0.772 nC14H30, 0.047 C14H30-2 and 0.181 C14H28-1. The simulations have been run for the engine speeds 900rpm, 1500rpm and 2500rpm, with altered fuel injections to approximate real engine conditions. To save computational time most of the cycles have been run from -20CAD to 60CAD, with inlet gas temperature and pressure of Ti=700K and Pi=2.33E06N/m2. The combustion characteristics of the different fuels and engine settings have been compared with regard to parameters such as pressure and temperature profiles, heat release, converted fuel and the production of the criteria pollutants CO, CO2 and NOx. Only small alterations in the combustion cycle is seen for the FTD surrogate with regard to n-heptane, which is most likely an effect of the substantial simplifications applied in the models. These simplifications were also evident in the last part of the thesis, where the emission profiles from the simulations were compared to experimental values. Here some of the cases showed deviating results, however others had correlating trends and could be qualitatively validated. The injection profiles were tuned with regard to the point of injection, where advanced injections obtain higher temperature and pressure peaks. Accordingly more work is produced, however also elevated levels of NOx. The effect of Exhaust Gas Recirculation (EGR) has been tested, where the NOx emissions are expected to diminish due to reduced concentration of oxygen. Here huge NOx reductions are observed, but also a trade-off with regard to elevated levels of CO and reduced levels of CO2. This reflects the main problem to be resolved when applying EGR, namely to what extent EGR should be applied before the negative effects related to a more incomplete combustion surpasses the NOx savings.