A study on flame reconstruction in a supersonic combustor using deep learning
- 1 January 2025
- journal article
- research article
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (1)
Abstract
This study investigates the application of a low-order reconstruction method for image reconstruction of a scramjet combustor. In the encoding network, reconstruction performance was assessed by evaluating adjustments to sampling channel count and modifications to neural network architectures. Upsampling methods such as convolutional neural networks (CNNs), interlayer attention mechanisms, and pixel shuffle were tested in the decoder network. Furthermore, a parameter expansion strategy based on the enlargement of convolutional feature map channels was proposed and examined. The results were quantified by morphological and frequency domain analyses under tests with datasets of different equivalence ratios, suggesting the effectiveness of the scheme for flashback prediction. It was found that the reconstruction effect of 6-point sampling is close to that of continuous sampling (68 points), which is the most cost-effective among the tested schemes. By comparing different network structures, the method proposed in this paper achieves better reconstruction results than the large-parameter CNN network with a small-scale network structure.Keywords
Funding Information
- National Natural Science Foundation of China (12102110)
- Advanced Jet Propulsion Innovation Center, AEAC (HKCX2022-02-007)
This publication has 33 references indexed in Scilit:
- Intelligent reconstruction algorithm of hydrogen-fueled scramjet combustor flow based on knowledge distillation model compressionInternational Journal of Hydrogen Energy, 2023
- Deep-learning prediction and uncertainty quantification for scramjet intake flowfieldsAerospace Science and Technology, 2022
- Sparse reconstruction of surface pressure coefficient based on compressed sensingExperiments in Fluids, 2022
- Prediction model of temperature field in dual-mode combustors based on wall pressureActa Astronautica, 2022
- Optimization problems for machine learning: A surveyEuropean Journal of Operational Research, 2021
- Monitoring combustion instabilities of stratified swirl flames by feature extractions of time-averaged flame images using deep learning methodAerospace Science and Technology, 2021
- Machine learning in additive manufacturing: State-of-the-art and perspectivesAdditive Manufacturing, 2020
- Analysis of combustion instability of hydrogen fueled scramjet combustor on high-speed OH-PLIF measurements and dynamic mode decompositionInternational Journal of Hydrogen Energy, 2020
- Flame oscillation characteristics in a kerosene fueled dual mode combustor equipped with thin strut flameholderActa Astronautica, 2019
- Investigation of flame establishment and stabilization mechanism in a kerosene fueled supersonic combustor equipped with a thin strutAerospace Science and Technology, 2017