Application of a Convolutional Neural Network for Wave Mode Identification in a Rotating Detonation Combustor Using High-Speed Imaging

Utilizing a neural network, individual down-axis images of combustion waves in the rotating detonation engine (RDE) can be classified according to the number of detonation waves present and their directional behavior. While the ability to identify the number of waves present within individual images might be intuitive, the further classification of wave rotational direction is a result of the detonation wave’s profile, which suggests its angular direction of movement. The application of deep learning is highly adaptive and, therefore, can be trained for a variety of image collection methods across RDE study platforms. In this study, a supervised approach is employed where a series of manually classified images is provided to a neural network for the purpose of optimizing the classification performance of the network. These images, referred to as the training set, are individually labeled as one of ten modes present in an experimental RDE. Possible classifications include deflagration, clockwise and counterclockwise variants of co-rotational detonation waves with quantities ranging from one to three waves, as well as single, double, and triple counter-rotating detonation waves. After training the network, a second set of manually classified images, referred to as the validation set, is used to evaluate the performance of the model. The ability to predict the detonation wave mode in a single image using a trained neural network substantially reduces computational complexity by circumnavigating the need to evaluate the temporal behavior of individual pixel regions throughout time. Results suggest that while image quality is critical, it is possible to accurately identify the modal behavior of detonation waves based on only a single image rather than a sequence of images or signal processing. Successful identification of wave behavior using image classification serves as a steppingstone for further machine learning integration in RDE research and development of comprehensive real-time diagnostics.

[1]  E. Gutmark,et al.  High-speed imaging of wave modes in an RDC , 2019, Experimental Thermal and Fluid Science.

[2]  Zenghui Wang,et al.  Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review , 2017, Neural Computation.

[3]  Ar Reed,et al.  An Intelligent Vision System for Monitoring and Control of Combustion Flames , 1999 .

[4]  Kristyn B. Johnson,et al.  Validation of Cross-Correlation Detonation Wave Mode Identification Through High-Speed Image Analysis , 2020 .

[5]  Douglas Schwer,et al.  Effect of Inlet on Fill Region and Performance of Rotating Detonation Engines , 2011 .

[6]  Michael T. Tong.  Using Machine Learning to Predict Core Sizes of High-Efficiency Turbofan Engines , 2019 .

[7]  G. Paniagua,et al.  Characterization of a Supersonic Turbine Downstream of a Rotating Detonation Combustor , 2018, Journal of Engineering for Gas Turbines and Power.

[8]  Vaibhav Joshi,et al.  Data-Driven Computing With Convolutional Neural Networks for Two-Phase Flows: Application to Wave-Structure Interaction , 2018, Volume 2: CFD and FSI.

[9]  Daniel E. Paxson,et al.  Foundational Performance Analyses of Pressure Gain Combustion Thermodynamic Benefits for Gas Turbines , 2012 .

[10]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[11]  Volker Sick,et al.  High speed imaging in fundamental and applied combustion research , 2013 .

[12]  S. Heister,et al.  Experimental Investigation of a Piloted, Natural Gas-Air Rotating Detonation Wave Combustor , 2018, 2018 Joint Propulsion Conference.

[13]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Toshio Nakahira,et al.  An image analysis of high speed combustion photographs for D.I. diesel engine with high pressure fuel injection , 1990 .

[15]  Hu Ma,et al.  Experimental study on a rotating detonation combustor with an axial-flow turbine , 2018, Acta Astronautica.

[16]  Min Xia,et al.  Fault Diagnosis for Rotating Machinery Using Multiple Sensors and Convolutional Neural Networks , 2018, IEEE/ASME Transactions on Mechatronics.

[17]  Stephen D. Heister,et al.  Parametric Survey of a Natural Gas-Air Rotating Detonation Engine at Elevated Pressure , 2019, AIAA Scitech 2019 Forum.

[18]  Christian Oliver Paschereit,et al.  Single and Counter-Rotating Wave Modes in an RDC , 2018 .

[19]  John Hoke,et al.  Imaging of OH* Chemiluminescence in an Optically Accessible Nonpremixed Rotating Detonation Engine , 2015 .

[20]  Joseph E. Shepherd,et al.  Detonation in gases , 2009 .

[21]  Matthew L. Dering,et al.  A Convolutional Neural Network Model for Predicting a Product's Function, Given Its Form , 2017 .

[22]  S. Heister,et al.  Performance Characterization of a Natural Gas-Air Rotating Detonation Engine at Elevated Pressure , 2019, AIAA Propulsion and Energy 2019 Forum.

[23]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[24]  E. Gutmark,et al.  Cross-correlation as a tool for measuring RDC wave speed, direction, and complexity , 2018, 2018 Joint Propulsion Conference.

[25]  John Hoke,et al.  T63 Turbine Response to Rotating Detonation Combustor Exhaust Flow , 2018, Journal of Engineering for Gas Turbines and Power.

[26]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[27]  Matthew Fotia,et al.  Experimental Performance Scaling of Rotating Detonation Engines Operated on Gaseous Fuels , 2017 .

[28]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[29]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[30]  Harry K. Moffat,et al.  Cantera: An Object-oriented Software Toolkit for Chemical Kinetics, Thermodynamics, and Transport Processes. Version 2.2.1 , 2016 .

[31]  W. Hargus,et al.  Automated image processing method to quantify rotating detonation wave behavior. , 2019, The Review of scientific instruments.

[32]  Elena Collado Morata,et al.  Thermodynamic analysis of a gas turbine engine with a rotating detonation combustor , 2017 .