Session: 04-33: Combustion Dynamics - Experiments III

Paper Number: 83006

83006 - System Parameter Identification of a Colored-Noise-Driven Rijke Tube Simulator 

Thermoacoustic oscillations in gas turbine combustors lead to sub-optimal performance, an increase in pollutant emissions and even catastrophic structural damage. Passive acoustic damping devices, such as Helmholtz resonators, are, commonly employed in modern combustors to avoid the occurrence of thermoacoustic oscillations. The estimation of deterministic parameters that characterize the thermoacoustic instability in a combustor are often required for designing these devices (Noiray and Schuermans, 2013). Because gas turbine combustors are typically noisy systems, it is important to understand the impact of noise on the thermoacoustic instability of the system. Most often, the reported works on noise-induced dynamics model the combustor noise as additive white Gaussian noise. However, experiments have shown that noise in combustors, in particular from the turbulent flame, is of a low-pass character (Bonciolini et al. 2016, 2017).
In this paper, we report an experimental investigation on the influence of colored noise (generated by the Ornstein-Uhlenbeck process) on the thermoacoustic coupling in an electroacoustic Rijke tube simulator (a prototypical thermoacoustic system). In the absence of noise, the system exhibits subcritical Hopf bifurcation and we study noise-induced response in the corresponding stable, bistable and linearly-unstable regimes. We use the Fokker-Planck formulation to investigate the effects of noise power and correlation time. We compare the system response when subjected to the colored noise and the white noise and find significant differences in the results. We find that for longer noise correlation time and higher noise power, the amplitude probability density function (PDF) from the two noise models sensibly differs which may lead to inaccurate parameter identification if a white noise model is used for the real system. This shows the importance of using accurate noise models for system identification of gas turbine systems.

Presenting Author: Neha Vishnoi Indian Institute of Technology, Ropar

Presenting Author Biography: Neha Vishnoi is a 3rd year PhD scholar in the Department of Mechanical Engineering at the Indian Institute of Technology Ropar, India. She is working on nonlinear and stochastic aspects of thermoacoustic instability in gas turbine combustors.

Authors:

Neha Vishnoi Indian Institute of Technology, Ropar
Vikrant Gupta Southern University of Science & Technology
Aditya Saurabh Indian Institute of Technology, Kanpur
Lipika Kabiraj Indian Institute of Technology, Ropar