|Prof. Dr.-Ing. Thomas Sattelmayer|| |
Thermoacoustic instabilities, acoustic liners, modeling
|Alireza Javareshkian, M.Sc.||This project is funded by the European Union Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement No 766264.|
In order to comply with the strict emission regulations, gas turbines are widely operated in lean premixed mode. In lean combustion, the device operates with excess air to reduce the combustion temperature and consequently formation of Nitrogen oxides. As a result, the device becomes highly prone to the pulsations. Therefore, it is desirable to increase acoustic damping in order to cope with the thermoacoustic instabilities.
In this project, acoustically absorbing liners are selected for further investigations. Therefore, the damping behavior of the absorbing liners will be characterized experimentally under cold flow and combustion regimes and will be compared against the combustor with rigid walls (without absorbing liner). Furthermore, in order to be capable of predicting the damping behavior of absorbing liners, a modeling approach is developed for the acoustic impedance of orifices on the wall (liner).
In order to predict the damping behavior of acoustic dampers, several acoustic impedance models based on characteristics of orifices are presented in the literature. One of the important aspects of modeling which is not carefully handled in the available models is the effect of interacting holes (hole-to-hole interaction). The so-called interaction effect was addressed in the literature as a consequence of increasing porosity (perforation rate) which affects the mass end correction by curtailing the attached mass contributing in oscillatory motion (depicted in Figure 1). According to this representation, for two interacting orifices, the mass end correction will be decreased by decreasing the distance of the orifices.
However, calculation of the impedance from pressure distribution around the orifice predicts the behavior of interacting orifices such that the mass end correction will be increased by decreasing the distance of the orifices which is shown in Figure 2 (Ingard ).
Nevertheless, the simplifying assumptions in the derivation process by Ingard , such as infinitely long tube and low frequency, prevent it from being reliably applicable to a wide range of applications. In addition to the modeling, further calibration of the model's parameters using Gaussian Processes will be carried out. For this purpose, the model's parameters will be calibrated based on the comparison of the model's predicted impedance and the measured impedance of orifices on the wall.
In order to observe the effect of interacting holes and using the experimental data for calibration of the impedance model, experiments with the so-called impedance tube are performed. Therefore, the acoustic impedance of the perforated plates and the associated backing cavity is measured. The test setup which is shown in Figure 3 is composed of an excitation source at the bottom, measuring module where the pressure sensors are mounted on and the absorbing module (perforated plate and backing cavity) on the top.
Regarding the current status of the project, the acoustic impedance of various perforated plates is measured and used as a benchmark for validation of the impedance model. Further calibration of the impedance model with the measured impedance will be carried out.
Moreover, for acoustic characterization, a suitable configuration of the absorbing liner will be manufactured and mounted on the combustor. Therefore, the acoustic damping rate in the presence of absorbing liner will be measured and compared with the combustor with rigid walls under cold flow and combustion regimes.
 U. Ingard, On the Theory and Design of Acoustic Resonators, Journal of the Acoustical Society of America, 25 (1953).