Combustion instability, or the coupling between acoustic oscillations and heat release rate fluctuations, is one of the most expensive problems facing power generation gas turbines today. The push to produce power with fewer emissions and increased cycling requirements brought on by alternative energy sources have made operating conditions prone to combustion instability more prevalent and in need of effective control measures. The potential solution for controlling these instabilities is active control systems, but a robust, active control system has not yet been developed for implementation on a full scale system.
Key to developing such a system is model parameter estimation. Without accurate model parameters, active control cannot be done. This thesis details the design, construction, and data acquisition capability of a modified Rijke tube built by the author for this investigation. The thermoacoustic behavior of the Rijke tube is mapped as a function of co-flow velocity and flame height. Sensitivity to system parameters is investigated. The results of the instability mapping and sensitivity analysis are used to inform the optimal experimental design. Finally, optimal experimental design for estimation of combustion model parameters of a linear time-delay thermoacoustic instability model is performed and results discussed.