Robust flame parameterization using image processing on PLIF experiments

The increasing demand of energy and the imposed emissions regulations call for energy conversion systems that are both environmentally friendly and as economically efficient as possible. For this purpose, natural gas is commonly used in combustion appliances as it often regarded as the cleanest of fossil fuels. The reduction of nitrous oxides (NOx) is particularly of importance, which can be achieved by lean combustion. Although this is beneficial for the NOx formation, it also causes destabilization of the flame.

Low-swirl stabilization is a promising new technology that achieves flame stabilization by a diverging flow field. This type of flow field lowers the residence time of gases at high temperature, causing a reduction of NOx emissions. However, low-swirl stabilization decreases the mixing near the flame resulting in limited conversion rates. In order to increase conversion rates, resonant turbulence is introduced to the flow.

The goal of this graduation project was to implement a robust method for flame parameterization, in order to obtain a method which can be used later in the project for further quantification of the effects imposed by resonant turbulence. First experiments on the low swirl flame were conducted using planar laser-induced fluorescence (PLiF) diagnostics, where after a robust processing procedure was designed and implemented capable of quantifying flame characteristics. The result is a method that is capable of processing PLiF material and quantifying flame parameters, even for low data quality.