0
Research Papers: Gas Turbines: Combustion, Fuels, and Emissions

Application of Artificial Neural Networks for Misfiring Detection in an Annular Pulsed Detonation Combustor Mockup

[+] Author and Article Information
Sascha Wolff

Chair of Measurement and Control,
Technische Universität Berlin,
Berlin 10623, Germany
e-mail: sascha.wolff@tu-berlin.de

Jan-Simon Schäpel

Chair of Measurement and Control,
Technische Universität Berlin,
Berlin 10623, Germany
e-mail: jan-simon.schaepel@tu-berlin.de

Rudibert King

Chair of Measurement and Control,
Technische Universität Berlin,
Berlin 10623, Germany
e-mail: rudibert.king@tu-berlin.de

1Corresponding author.

Contributed by the Combustion and Fuels Committee of ASME for publication in the JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript received July 15, 2016; final manuscript received August 24, 2016; published online November 16, 2016. Editor: David Wisler.

J. Eng. Gas Turbines Power 139(4), 041510 (Nov 16, 2016) (7 pages) Paper No: GTP-16-1342; doi: 10.1115/1.4034941 History: Received July 15, 2016; Revised August 24, 2016

An annular pulsed detonation combustor (PDC) basically consists of a number of detonation tubes which are firing in a predetermined sequence into a common downstream annular plenum. Fluctuating initial conditions and fluctuating environmental parameters strongly affect the detonation. Operating such a setup without misfiring is delicate. Misfiring of individual combustion tubes will significantly lower performance or even stop the engine. Hence, an operation of such an engine requires a misfiring detection. Here, a supervised data driven machine learning approach is used for the misfiring detection. The features used as inputs for the classifier are extracted from measurements incorporating physical knowledge about the given setup. To this end, a neural network is trained based on labeled data which is then used for classification purposes, i.e., misfiring detection. A surrogate, nonreacting experimental setup is considered in order to develop and test these methods.

FIGURES IN THIS ARTICLE
<>
Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.

References

Roy, G. , Frolov, S. , Borisov, A. , and Netzer, D. , 2004, “ Pulse Detonation Propulsion: Challenges, Current Status, and Future Perspective,” Prog. Energy Combust. Sci., 30(6), pp. 545–672. [CrossRef]
Bobusch, B. C. , Berndt, P. , Paschereit, C. O. , and Klein, R. , 2014, “ Shockless Explosion Combustion: An Innovative Way of Efficient Constant Volume Combustion in Gas Turbines,” Combust. Sci. Technol., 186(10–11), pp. 1680–1689. [CrossRef]
Wolff, S. , and King, R. , 2015, “ Model-Based Detection of Misfirings in an Annular Burner Mockup,” Active Flow and Combustion Control III (NNFM, Vol. 127), R. King , ed., Springer, Heidelberg, Germany, pp. 229–244.
Wolff, S. , and King, R. , 2015, “ An Annular Pulsed Detonation Combustor Mockup: System Identification and Misfiring Detection,” ASME J. Eng. Gas Turbines Power, 138(4), p. 041603. [CrossRef]
Chandola, V. , Banerjee, A. , and Kumar, V. , 2009, “ Anomaly Detection: A Survey,” ACM Comput. Surv., 41(3), pp. 15:1–15:58. [CrossRef]
Allegoric, C. , and Mantini, V. , 2014, “ A Data-Driven Approach for On-Line Gas Turbine Combustion Monitoring Using Classification Models,” Second European Conference of the Prognostics and Health Management Society 2014, pp. 92–100.
Romesis, C. , and Mathioudakis, K. , 2003, “ Setting Up of a Probabilistic Neural Network for Sensor Fault Detection Including Operation With Component Faults,” ASME J. Eng. Gas Turbines Power, 125(3), pp. 634–641. [CrossRef]
Dreyfus, G. , 2010, Neural Networks: Methodology and Applications, Springer, Heidelberg, Germany.
Hagan, M. T. , Demuth, H. B. , Beale, M. H. , and de Jesús, O. , 2014, Neural Network Design, 2nd ed., Oklahoma State University, Stillwater, OK.

Figures

Grahic Jump Location
Fig. 1

Annular gap with 12 circular holes in the front plate (at z = 0) for the connection of surrogate firing tubes/actuators. The dimensions of the system are given by L = 0.6 m, aR = 0.32 m, R = 0.4 m, and a = 0.8.

Grahic Jump Location
Fig. 2

Experimental setup with DSP, amplifiers, actuators, and sensors

Grahic Jump Location
Fig. 3

Propagation of acoustic waves from loudspeakers through the plenum. Loudspeaker one and two generate the signals f1(t) and f2(t), respectively.

Grahic Jump Location
Fig. 4

Model of a single neuron

Grahic Jump Location
Fig. 5

Exemplified structure of an artificial neural network with input, hidden, and output layer

Grahic Jump Location
Fig. 6

Misfiring detection results for the shown misfiring cases. The percentage of detected misfirings for each tube is depicted.

Grahic Jump Location
Fig. 7

Detection quality of misfiring patterns where only single tubes misfire

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In