Desenvolvimento de métodos computacionais para análise da injeção direta de combustíveis líquidos

Abstract

The analysis of better fuel applications is of great importance when is considered the growing restrictions on fossil fuel emissions, and its environmental implications in the actual context. Studying the fuel in the way it is used is a strategy for reaching this objective. This work has the goal to improve the techniques of analysis concerning some relevant characteristics of direct injection using liquid fuels in internal combustion engines. In this work computational programming calculations were linked with experimental data for getting more reliable results and mathematical relations for spray relevant parameters. Images are the source for getting results concerning fuel spray analysis in several experiments found in the literature. In the present work images were also used for this objective. They are essentially numerical matrices in a mathematical representation. High-speed image recording processes were performed both in static injection conditions using gasoline and ethanol similar fluids and in dynamic conditions with the ethanol injections inside an optical engine to serve as an example for the application of the new methodologies. With the developed strategies, it was possible to measure automatically important spray parameters as the cone angle, the penetration, the spray volume and the dispersion coefficients. Such strategies also helped obtaining predictive correlations. The results showed similar behavior in the macroscopic spray characteristics between the two tested fluids used in static conditions and a strong dependence in ethanol spray with the engine speed rotation.