REFINED STUDY OF SUPER-RESOLUTION RECONSTRUCTION OF NEAR-WALL TURBULENCE FIELD BASED ON CNN AND GAN DEEP LEARNING MODEL

Wu Haokai; Chen Yaoran; Zhou Dai; Chen Wenli; Cao Yong

doi:10.6052/0459-1879-24-019

Volume 56 Issue 8

Aug. 2024

Turn off MathJax

Article Contents

Abstract

References

Chinese Journal of Theoretical and Applied Mechanics > 2024 > 56(8): 2231-2242. > DOI: 10.6052/0459-1879-24-019 CSTR: 32045.14.0459-1879-24-019

Wu Haokai, Chen Yaoran, Zhou Dai, Chen Wenli, Cao Yong. Refined study of super-resolution reconstruction of near-wall turbulence field based on CNN and GAN deep learning model. Chinese Journal of Theoretical and Applied Mechanics, 2024, 56(8): 2231-2242. DOI: 10.6052/0459-1879-24-019

Citation:

PDF (3640 KB)

REFINED STUDY OF SUPER-RESOLUTION RECONSTRUCTION OF NEAR-WALL TURBULENCE FIELD BASED ON CNN AND GAN DEEP LEARNING MODEL

Wu Haokai^*,
Chen Yaoran^†,
Zhou Dai^*,
Chen Wenli^**,
Cao Yong^{*, ††, ,}

*.
School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
†.
School of Future Technology, Shanghai University, Shanghai 200444, China
**.
School of Civil Engineering, Harbin Institute of Technology, Harbin 150001, China
††.
Chongqing Research Institute, Shanghai Jiao Tong University, Chongqing 400799, China

Received Date: January 04, 2024
Accepted Date: March 09, 2024
Available Online: March 09, 2024
Published Date: March 10, 2024

Graphical Abstract

Abstract

Abstract

For the purpose of urban wind resistance and disaster reduction, the reconstruction of urban boundary layer with high fidelity is a key issue to be solved in the wind engineering. Based on high-resolution near-ground wind field, it is expected to achieve the accurate prediction of the wind-induced effect of urban buildings in the real environment. Traditional simulation method based on meteorological models have shortcomings including long forecasting time, high computational cost and limited solution resolution. In order to predict the spatial variation of the near-wall turbulence more accurately and efficiently, the super-resolution convolutional neural network (SRCNN) and generative adversarial neural network (SRGAN) are applied to reconstruct super-resolution near-wall turbulence field from the low-resolution one. This study employs the public database, which is built up from direct numerical simulation of turbulent channel flow, to train the reconstruction models and evaluate their performance. In order to determine a suitable model generation strategy, this study searches for a proper training neural network and its optimal parameter setting based on detailed sensitivity analysis of the training sample size and network depth. What’s more, the application scope of the model is explored for near-wall flow field reconstruction, based on low-resolution inputs obtained by different down scale ratios. It is found that the SRGAN model has a stronger capacity to reproduce the small-scale structures in the turbulent near-wall flow, compared with the SRCNN model. As the training is based on 300 sample sizes and 4 convolutional residual blocks, the generated SRGAN model can obtain a good reconstruction accuracy, at a relatively lower training cost. When 10 times super-resolution reconstruction is carried out, the SRGAN model can still maintain ideal prediction performance. The research results offer reliable technical support for reconstructing near-wall turbulence based on artificial intelligence. Subsequently, it provides precise inflow conditions to efficiently predict the wind-induced effects on buildings in urban areas.

FullText(HTML)

References (30)

References

[1]	刘天绍, 刘孙俊, 杨玺等. 1951-2015影响广东沿海台风的统计分析. 海洋预报, 2018, 35(4): 68-74 (Liu Tianshao, Liu Sunjun, Yang Xi, et al. Statistical analysis of the typhoon influencing Guangdong province during 1951-2015. Marine Forecasts, 2018, 35(4): 68-74 (in Chinese) Liu Tianshao, Liu Sunjun, Yang Xi, et al. Statistical analysis of the typhoon influencing Guangdong province during 1951-2015. Marine Forecasts, 2018, 35(4): 68-74 (in Chinese)
[2]	雷鹰, 李涛, 张建国等. 厦门地区台风风场特性的数值模拟. 工程力学, 2014, 31(1): 122-128 (Lei Ying, Li Tao, Zhang Jianguo, et al. Numerical simulation of the characteristics of typhoon wind-field in Xiamen region. Engineering Mechanics, 2014, 31(1): 122-128 (in Chinese) Lei Ying, Li Tao, Zhang Jianguo, et al. Numerical simulation of the characteristics of typhoon wind-field in Xiamen region. Engineering Mechanics, 2014, 31(1): 122-128 (in Chinese)
[3]	杨剑. 基于物理过程的台风多灾害模拟与危险性分析. [博士论文]. 哈尔滨: 哈尔滨工业大学, 2022 (Yang Jian. Physics based typhoon multi-hazard modeling and hazard analysis. [PhD Thesis]. Harbin: Harbin Institute of Technology, 2022 (in Chinese) Yang Jian. Physics based typhoon multi-hazard modeling and hazard analysis. [PhD Thesis]. Harbin: Harbin Institute of Technology, 2022 (in Chinese)
[4]	王辉, 吴亚雄, 吴学健等. 低矮异型建筑立面洞口对下击暴流风压作用的影响研究. 合肥工业大学学报(自然科学版), 2023, 46(11): 1484-1491 (Wang Hui, Wu Yaxiong, Wu Xuejian, et al. Study on the effect of facade openings on wind pressure for low-rise irregular buildings under downburst. Journal of Hefei University of Technology (Natural Science), 2023, 46(11): 1484-1491 (in Chinese) Wang Hui, Wu Yaxiong, Wu Xuejian, et al. Study on the effect of facade openings on wind pressure for low-rise irregular buildings under downburst. Journal of Hefei University of Technology (Natural Science), 2023, 46(11): 1484-1491 (in Chinese)
[5]	Luo YP, Fu JY, Li QS, et al. Observation of Typhoon Hato based on the 356-m high meteorological gradient tower at Shenzhen. Journal of Wind Engineering and Industrial Aerodynamics, 2020, 207: 104408 doi: 10.1016/j.jweia.2020.104408
[6]	Xue L, Li Y, Song L, et al. A WRF-based engineering wind field model for tropical cyclones and its applications. Natural Hazards, 2017, 87(3): 1735-1750 doi: 10.1007/s11069-017-2845-z
[7]	Tan C, Fang W. Mapping the wind hazard of global tropical cyclones with parametric wind field models by considering the effects of local factors. International Journal of Disaster Risk Science, 2018, 9(1): 86-99 doi: 10.1007/s13753-018-0161-1
[8]	Tamura T. Towards practical use of LES in wind engineering. Journal of Wind Engineering and Industrial Aerodynamics, 2008, 96(10-11): 1451-1471 doi: 10.1016/j.jweia.2008.02.034
[9]	Takemi T, Yoshida T, Horiguchi M, et al. Large-eddy-simulation analysis of airflows and strong wind hazards in urban areas. Urban Climate, 2020, 32: 100625 doi: 10.1016/j.uclim.2020.100625
[10]	Brunton SL, Noack BR, Koumoutsakos P. Machine learning for fluid mechanics. Annual Review of Fluid Mechanics, 2020, 52(1): 477-508 doi: 10.1146/annurev-fluid-010719-060214
[11]	张伟伟, 王旭, 寇家庆. 面向流体力学的多范式融合研究展望. 力学进展, 2023, 53(2): 433-467 (Zhang Weiwei, Wang Xu, Kou Jiaqing. Prospects of multi-paradigm fusion methods for fluid mechanics research. Advances in Mechanics, 2023, 53(2): 433-467 (in Chinese) Zhang Weiwei, Wang Xu, Kou Jiaqing. Prospects of multi-paradigm fusion methods for fluid mechanics research. Advances in Mechanics, 2023, 53(2): 433-467 (in Chinese)
[12]	Duraisamy K, Xiao GIH. Turbulence modeling in the age of data. Annual Review of Fluid Mechanics, 2019, 51(1): 357-377 doi: 10.1146/annurev-fluid-010518-040547
[13]	胡震宇, 王子路, 陈坚强等. 基于深度神经网络的横流转捩预测. 力学学报, 2023, 55(1): 433-467 (Hu Zhenyu, Wang Zilu, Chen Jianqiang, et al. Prediction of crossflow transition based on deep neural networks. Chinese Journal of Theoretical and Applied Mechanics, 2023, 55(1): 433-467 (in Chinese) Hu Zhenyu, Wang Zilu, Chen Jianqiang, et al. Prediction of crossflow transition based on deep neural networks. Chinese Journal of Theoretical and Applied Mechanics, 2023, 55(1): 433-467 (in Chinese)
[14]	Srinivasan PA, Guastoni L, Azizpour H, et al. Predictions of turbulent shear flows using deep neural networks. Physical Review Fluids, 2019, 4(5): 054603 doi: 10.1103/PhysRevFluids.4.054603
[15]	郭振东, 成辉, 陈云等. 融合相似性原理的涡轮叶型流场预测方法研究. 力学学报, 2023, 55(11): 2647-2660 (Guo Zhendong, Cheng Hui, Chen Yun, et al. Study of flow field prediction of turbine blades by coupling similarity principle. Chinese Journal of Theoretical and Applied Mechanics, 2023, 55(11): 2647-2660 (in Chinese) Guo Zhendong, Cheng Hui, Chen Yun, et al. Study of flow field prediction of turbine blades by coupling similarity principle. Chinese Journal of Theoretical and Applied Mechanics, 2023, 55(11): 2647-2660 (in Chinese)
[16]	余柏杨, 吕宏强, 周岩等. 基于机器学习的高速复杂流场流动控制效果预测分析. 实验流体力学, 2022, 36(3): 44-54 (Yu Baiyang, Lv Hongqiang, Zhou Yan, et al. Predictive analysis of flow control in high-speed complex flow field based on machine learning. Journal of Experiments in Fluid Mechanics, 2022, 36(3): 44-54 (in Chinese) Yu Baiyang, Lv Hongqiang, Zhou Yan, et al. Predictive analysis of flow control in high-speed complex flow field based on machine learning. Journal of Experiments in Fluid Mechanics, 2022, 36(3): 44-54 (in Chinese)
[17]	Fan D, Yang L, Wang Z, et al. Reinforcement learning for bluff body active flow control in experiments and simulations. Proceedings of the National Academy of Sciences, 2020, 117(42): 26091-26098 doi: 10.1073/pnas.2004939117
[18]	林子文. 基于深度学习的海洋时空数据预测技术研究. [硕士论文]. 哈尔滨: 哈尔滨工业大学, 2022 (Lin Ziwen. Research on marine spatiotemporal data prediction technology based on deep learning. [Master Thesis]. Harbin: Harbin Institute of Technology, 2022 (in Chinese) Lin Ziwen. Research on marine spatiotemporal data prediction technology based on deep learning. [Master Thesis]. Harbin: Harbin Institute of Technology, 2022 (in Chinese)
[19]	Fukami K, Fukagata K, Taira K. Super-resolution reconstruction of turbulent flows with machine learning. Journal of Fluid Mechanics, 2019, 870: 106-120 doi: 10.1017/jfm.2019.238
[20]	Fukami K, Fukagata K, Taira K. Machine learning-based spatio-temporal super resolution reconstruction of turbulent flows. Journal of Fluid Mechanics, 2021, 909: A9 doi: 10.1017/jfm.2020.948
[21]	Cai S, Zhou S, Xu C, et al. Dense motion estimation of particle images via a convolutional neural network. Experiments in Fluids, 2019, 60: 73 doi: 10.1007/s00348-019-2717-2
[22]	Wu H, Chen Y, Xie P, et al. Sparse-measurement-based peak wind pressure evaluation by super-resolution convolutional neural networks. Journal of Wind Engineering & Industrial Aerodynamics, 2023, 242: 105574
[23]	Güemes A, Discetti S, Ianiro A, et al. From coarse wall measurements to turbulent velocity fields through deep learning. Physics of Fluids, 2021, 33: 075121 doi: 10.1063/5.0058346
[24]	Deng Z, He C, Liu Y, et al. Super-resolution reconstruction of turbulent velocity fields using a generative adversarial network-based artificial intelligence framework. Physics of Fluids, 2019, 31(12): 125111 doi: 10.1063/1.5127031
[25]	Yousif MZ, Yu L, Lim H, et al. Super-resolution reconstruction of turbulent flow fields at various Reynolds numbers based on generative adversarial networks. Physics of Fluids, 2022, 34: 015130 doi: 10.1063/5.0074724
[26]	Fukami K, Nabae Y, Kawai K, et al. Synthetic turbulent inflow generator using machine learning. Physical Review Fluids, 2019, 4(6): 064603
[27]	Yousif MZ, Yu L, Lim H. Physics-guided deep learning for generating turbulent inflow conditions. Journal of Fluid Mechanics, 2022, 936: A21
[28]	Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets. Advances in Neural Information Processing Systems, 2014, 27: 2672-2680
[29]	Ledig C, Theis L, Huszar F, et al. Photo-realistic single image super-resolution using a generative adversarial network//IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, 2017
[30]	He K, Zhang X, Ren S, et al. Residual learning for image recognition//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, USA, 2016

[1]	Luo Renyu, Li Qizhi, Zu Gongbo, Huang Yunjin, Yang Gengchao, Yao Qinghe. A SUPER-RESOLUTION LATTICE BOLTZMANN METHOD BASED ON CONVOLUTIONAL NEURAL NETWORK[J]. Chinese Journal of Theoretical and Applied Mechanics, 2024, 56(12): 3612-3624. DOI: 10.6052/0459-1879-24-248
[2]	Chen Hao, Guo Mingming, Tian Ye, Chen Erda, Deng Xue, Le Jialing, Li Linjing. PROGRESS OF CONVOLUTION NEURAL NETWORKS IN FLOW FIELD RECONSTRUCTION[J]. Chinese Journal of Theoretical and Applied Mechanics, 2022, 54(9): 2343-2360. DOI: 10.6052/0459-1879-22-130
[3]	Gao Jun, Li Jia. NUMERICAL INVERSITAGION OF MODE EXCHANGE IN HYPERSONIC BOUNDARY LAYERS[J]. Chinese Journal of Theoretical and Applied Mechanics, 2018, 50(6): 1368-1378. DOI: 10.6052/0459-1879-18-260
[4]	Gan Yangke, Liu Jianfei. AUTOMATIC VISCOUS BOUNDARY LAYER MESH GENERATION[J]. Chinese Journal of Theoretical and Applied Mechanics, 2017, 49(5): 1029-1041. DOI: 10.6052/0459-1879-17-154
[5]	Liang Tinghao, Yu Xiping. A NUMERICAL STUDY ON THERMAL INTERNAL BOUNDARY LAYER OVER A COASTAL CITY[J]. Chinese Journal of Theoretical and Applied Mechanics, 2016, 48(2): 473-481. DOI: 10.6052/0459-1879-15-056
[6]	The research of plume dispersion in the atmosphere boundary layer[J]. Chinese Journal of Theoretical and Applied Mechanics, 2005, 37(2): 148-156. DOI: 10.6052/0459-1879-2005-2-2004-187
[7]	Flow control effects of electromagnetic force in the boundary layer[J]. Chinese Journal of Theoretical and Applied Mechanics, 2004, 36(4): 472-478. DOI: 10.6052/0459-1879-2004-4-2003-392
[8]	AN INVESTIGATION OF SURFACE BOUNDARY CONDITIONS FOR CHEMICALLY NON EQUILIBRIUM BOUNDABY LAYERS[J]. Chinese Journal of Theoretical and Applied Mechanics, 1994, 26(4): 407-415. DOI: 10.6052/0459-1879-1994-4-1995-562
[9]	STEP INDUCED HYPERSONIC TURBULENT BOUNDARY-LAYER SEPARATION[J]. Chinese Journal of Theoretical and Applied Mechanics, 1994, 26(1): 113-120. DOI: 10.6052/0459-1879-1994-1-1995-528
[10]	THE WAYS OF THE FORMATION OF THE HORSESHOE VORTEX IN TURBULENT BOUNDARY LAYER[J]. Chinese Journal of Theoretical and Applied Mechanics, 1992, 24(2): 145-151. DOI: 10.6052/0459-1879-1992-2-1995-722

Cited By

Cited by

Periodical cited type(2)

1.	王瑄，孔辰，韩云霄，李佳，常军涛. 引入物理约束的航空发动机燃烧室温度场预测模型. 推进技术. 2024(12): 64-78 .
2.	罗仁宇，李奇志，祖公博，黄云进，杨耿超，姚清河. 基于卷积神经网络的超分辨率格子Boltzmann方法研究. 力学学报. 2024(12): 3612-3624 . 本站查看

Other cited types(0)

Get Citation

PDF

XML

Article Metrics

Article views (392) PDF downloads (120) Cited by(2)

REFINED STUDY OF SUPER-RESOLUTION RECONSTRUCTION OF NEAR-WALL TURBULENCE FIELD BASED ON CNN AND GAN DEEP LEARNING MODEL

Abstract

References

Related Articles

Cited by

Periodical cited type(2)

Other cited types(0)

Catalog

Article Metrics

Related

Download Center

Author Center

REFINED STUDY OF SUPER-RESOLUTION RECONSTRUCTION OF NEAR-WALL TURBULENCE FIELD BASED ON CNN AND GAN DEEP LEARNING MODEL

Abstract

References

Related Articles

Cited by

Periodical cited type(2)

Other cited types(0)

Catalog

Article Metrics

Related

Download Center

Author Center

Export File

Citation

Format

Content