AIP Advances
(Nov 2022)
Noise suppression method for hydroxyl tagging velocimetry based on generative adversarial networks
Jun Shao,
Junzheng Wu,
Jingfeng Ye,
Zhenjie Wu,
Zhenrong Zhang,
Sheng Wang,
Guohua Li,
Mengmeng Tao,
Haolong Wu,
Aiping Yi,
Zhiyun Hu
Affiliations
Jun Shao
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Junzheng Wu
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Jingfeng Ye
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Zhenjie Wu
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Zhenrong Zhang
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Sheng Wang
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Guohua Li
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Mengmeng Tao
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Haolong Wu
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Aiping Yi
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
Zhiyun Hu
State Key Laboratory of Laser Interaction with Matter, Northwest Institute of Nuclear Technology, Xi’an, Shaanxi 710024, People’s Republic of China
DOI
https://doi.org/10.1063/5.0121343
Journal volume & issue
Vol. 12,
no. 11
Abstract
Read online
Hydroxyl tagging velocimetry (HTV) technology is crucial in the velocimetry diagnosis of combustion flow fields. However, obtaining accurate HTV information in practical engineering applications is difficult because of complex flow fields and background noise interference. Therefore, for noise suppression, we proposed a generative adversarial network method for targeted network training based on the analysis of HTV image noise characteristics in a complex flow field and the construction of a high-confidence noise description model. The proposed method can effectively suppress noise in HTV experimental data, improve the signal-to-noise ratio of HTV images, and improve the accuracy of HTV measurement.
WeChat QR code
Close
