The application of laser diagnostics, light emitting diodes and computer modeling to the characterization of fuel-air mixing in high pressure lean premixed combustion systems

Minimizing the emissions of pollutants from combustion devices such as internal combustion engines and gas turbines is a challenge faced by the combustion engineer. In lean premixed combustion systems, such as homogeneous charge compression ignition engines or lean premixed gas turbines, sufficient mixing of the fuel and air prior to combustion can cause reductions in pollutant emissions.; Measuring the extent of mixing of fuel into air is often difficult, since combustion in these systems often takes place at very high pressures, optical access to the combustion area is often limited.; This thesis studies some of the problems in determining the extent of mixing of the fuel into air in these lean premixed combustion systems. The focus of this thesis is on three major topics: (1) the application of a genetic algorithm to the problem of tomographically reconstructing the mean (extensively studied) and RMS (a new result) of the fuel concentration field, (2) the use of an extractive laser probe for time-resolved fuel concentration measurements from gas turbines and internal combustion engines and (3) the use of an infrared light emitting diode (IR-LED) to quantitatively measure fuel concentration in a lean premixed gas turbine.; The use of tomographic reconstruction for generating the concentration field from a set of line of sight (LOS) absorption measurements is a fairly well known problem. Tomographic reconstruction is commonly called a CAT scan in the medical profession (Hounsfield and Cormack shared the Nobel Prize for Medicine for the development of Computer Assisted Tomography in 1979). However, much less is known about the ability to generate the RMS of the fuel concentration field from the RMS values of the LOS absorption measurements. This thesis examines the problem of RMS reconstruction, develops equations that relate the RMS of the LOS measurements to the RMS of the concentration field, and tests the method. This thesis shows that the relation between time-dependent LOS measurements and the RMS of the concentration field cannot be solved directly without some assumption about the spatial correlation of the time-dependent fuel concentration values. Thus, this thesis places bounds on the maximum and minimum values of the RMS of the concentration based on the two extreme cases: one where the spatial variations of the RMS of the concentration field are completely correlated, and the other where the spatial variations of the RMS of the concentration field have no correlation. (Abstract shortened by UMI.)...