Early Detection of Combustion Instability from Hi-speed Flame Images via Deep Learning and Symbolic Time Series Analysis

Soumalya Sarkar, Kin G. Lore, Soumik Sarkar, Vikram Ramanan, Satyanarayanan R Chakravarthy, Shashi Phoha, and Asok Ray
Submission Type: 
Full Paper
AttachmentSizeTimestamp
phmc_15_057.pdf658.88 KBAugust 24, 2015 - 4:47pm

Combustion instability has many detrimental effects on flight propulsion dynamics and structural integrity of gas turbine engines and its early detection is one of the important tasks in engine health monitoring and prognostics. Combustion instability is characterized by self-sustained, large-amplitude pressure oscillations and periodic shedding of coherent vortex structures at varied spatial scales. It is caused when localized hydrodynamic perturbations in fluid flow fluctuations are augmented by heat release, coupled with acoustics of the combustion chamber. This paper proposes a dynamic data driven approach, where a large volume of sequential hi-speed (greyscale) images is used to analyze the temporal evolution of coherent structures in combustion chamber for early detection of combustion instability at various operating conditions. The lower layer of the proposed hierarchical approach extracts low-dimensional semantic features from images using Deep Neural Networks. The upper layer captures the temporal evolution of the extracted features with a probabilistic graphical modeling scheme called Symbolic Time Series Analysis (STSA). Extensive experimental data have been collected in a swirl stabilized dump combustor at various operating conditions (e.g., premixing level and flow velocity) for validation of the proposed approach. Intermediate layer visualization of deep learning reveals that meaningful shape-features from the flame images are extracted, which enables the temporal modeling layer to enhance the class separability between stable and unstable regions. At the same time, the semantic nature of intermediate features enables expert guided data exploration that can lead to better understanding of the underlying physics. To the best of the authors knowledge, this paper presents one of the early applications of the recently reported Deep Learning tools in the area of prognostics and health management (PHM).

Publication Year: 
2015
Publication Volume: 
6
Publication Control Number: 
057
Page Count: 
10
Submission Keywords: 
Combustion instability
deep learning
time series analysis
probabilistic graphical model
Submission Topic Areas: 
Data-driven methods for fault detection, diagnosis, and prognosis
Submitted by: 
  
 
 
 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed