DATE : 21.04.2017
TIME : 4:00 PM
VENUE : ESB 244
SPEAKER : Suma.H (EE15D036)
GUIDE : Dr. Balaji Srinivasan
Dr. Anil Prabhakar (Chairperson)
Dr. Harishankar Ramachandran (M)
Dr. Boby George (M)
Dr. Raghavendra Sai.V.V (M)(AM)
Combustion instability poses a key limitation on the performance and structural durability of both land-based and aircraft gas turbine engines. As such, detection of the onset of combustion instability is not only critical for performance monitoring and fault diagnosis, but also for initiating efficient decision and control of such engines. Recently, a dynamic data driven method based on symbolic time series analysis (STSA) approach built upon the generalized D-Markov machine has been proposed for detecting early onset of thermo-acoustic instability in swirl-stabilized combustors. This approach models spatio-temporal co-dependence among time series from heterogeneous (e.g. pressure and chemi-luminescence) sensors to generate a robust data-driven precursor which is uniformly applicable across multiple experiment protocols with various premixing levels.
In order to experimentally validate the above approach, we propose a method wherein a fiber optic bundle can be used as the chemi-luminescence sensor to visualize the flame and fiber Bragg gratings can be used to monitor the acoustic emission inside the combustor. Since the combustion instability is a result of coupling between heat release and acoustics of the combustion chamber, the proposed sensor solution is expected to give optimum results. Results of the initial controlled experiments using fiber optic bundle and FBGs are also presented.
All are cordially invited.