Combining Machine Learning, Embedded Sensor Networks and Additive Burner Design for Combustor Structural Health Monitoring
Authors:
- Fabrice GiulianiFabrice Giuliani
- Nina PaulitschNPNina Paulitsch
- Andrea HoferAHAndrea Hofer
- Vojislav Petrovic-FilipovicVPVojislav Petrovic-Filipovic
- Benjamin MeierBMBenjamin Meier
- Werner BailerWBWerner Bailer
- Martin WinterMWMartin Winter
- Roland UnterbergerRURoland Unterberger
- Alexander SchrickerASAlexander Schricker
- 4 October 2024
- journal article
- research article
- Published by ASME International in Journal of Engineering for Gas Turbines and Power
- Vol. 147 (3) , 1-11
Abstract
The crystAIr project is about sensitizing burners for flame monitoring based on a discrete number of dynamic pressure sensors and a machine learning approach. The context is the development of better aircraft engines and the prospect of technical solutions for hydrogen-fuelled gas turbines. The combination of sensors and appropriate signal processing is designed to mimic a “feel” for the state of the flame. The idea is to monitor the gas turbine's correct operation in real-time and anticipate impending problems such as flame blowout, combustion instability or flashback. Compared to conventional flame monitoring based on signal sampling, thresholding, bandpass analysis and time-lag measurement, this method should be less computationally intensive, more sensitive and more robust to artifacts. Not only should it be more responsive, but it should also anticipate problems based on a lessons-learned approach and outperform the conventional precursors. An unassisted machine learning method is used to achieve this. A custom-built AM burner designed for this experiment provides multiple instrumentation options. A powerful data acquisition system is setup to collect data on multiple channels, over a long time duration and at high speed to collect the learning signal chunks. The focus is on the detection of flames and, more specifically, the moment of their ignition and extinction, the operating conditions that are prone to flashback and the stability of the combustion. Additional measurement techniques are used to refine the learning method. The paper covers all these points and presents the first results.Keywords
This publication has 9 references indexed in Scilit:
- Bayesian Machine Learning for the Prognosis of Combustion Instabilities From NoiseJournal of Engineering for Gas Turbines and Power, 2021
- A review on recent progress in computational and empirical studies of compression ignition internal combustion engineFuel, 2020
- Internal surface roughness enhancement of parts made by laser powder-bed fusion additive manufacturingVacuum, 2020
- Densification and microstructural investigation of Inconel 718 parts fabricated by selective laser meltingPowder Technology, 2017
- Microstructure and hardness studies of Inconel 718 manufactured by selective laser melting before and after solution heat treatmentMaterials Science and Engineering: A, 2017
- Short Helical Combustor: Concept Study of an Innovative Gas Turbine Combustor With Angular Air SupplyJournal of Engineering for Gas Turbines and Power, 2015
- Experimental and numerical study of MILD combustion for gas turbine applicationsApplied Energy, 2015
- Self-induced combustion oscillations of laminar premixed flames stabilized on annular burnersCombustion and Flame, 2003
- Application of Active Combustion Instability Control to a Heavy Duty Gas TurbineJournal of Engineering for Gas Turbines and Power, 1998