A robust super-resolution reconstruction model of turbulent flow data based on deep learning

doi:10.1016/j.compfluid.2022.105382

IMECH-IR > 非线性力学国家重点实验室

	A robust super-resolution reconstruction model of turbulent flow data based on deep learning
	Zhou ZD(周志登); Li BL(李秉霖); Yang XL(杨晓雷); Yang ZX(杨子轩)1
通讯作者	Yang, Zixuan(yangzx@imech.ac.cn)
发表期刊	COMPUTERS & FLUIDS
	2022-05-15
卷号	239 页码:15
ISSN	0045-7930
摘要	A new super-resolution model, namely the turbulence volumetric super-resolution (TVSR) model, is developed based on convolutional neural network (CNN) to reconstruct three-dimensional high-resolution turbulent flow field data from low-resolution data. Direct numerical simulation (DNS) and corresponding filtered DNS (FDNS) data of homogeneous isotropic turbulence at various Reynolds numbers are used to train the TVSR model. The proposed model is a modification of Liu et al. (2020), aiming to provide an improved generalization capability of the super-resolution model. For this purpose, we propose a patchwise training strategy in consideration of the property of turbulence that the velocity correlation between two points diminishes as the separation becomes sufficiently large. Furthermore, data at various Reynolds numbers are combined together to train the model. In comparison with existing models, the present TVSR model shows a better generalization capability in two aspects. First, the TVSR model trained using data at low Reynolds numbers is found robust and accurate in the super-resolution reconstructions of flow fields at higher Reynolds numbers. Second, although only DNS data are used for training, the TVSR model is also robust in reconstructing high-resolution flow fields from low-resolution data obtained from large-eddy simulation (LES). This feature of the TVSR model provides a new access to obtain turbulent motions at unresolved scales in LES studies of turbulent flows.
关键词	Super-resolution model Direct numerical simulation Large-Eddy simulation Isotropic turbulence Unresolved scales
DOI	10.1016/j.compfluid.2022.105382
收录类别	SCI ; EI
语种	英语
WOS记录号	WOS:000793060900001
关键词[WOS]	LARGE-EDDY SIMULATIONS ; ISOTROPIC TURBULENCE ; TIME CORRELATIONS ; DECONVOLUTION ; ENRICHMENT ; SCALES
WOS研究方向	Computer Science ; Mechanics
WOS类目	Computer Science, Interdisciplinary Applications ; Mechanics
资助项目	National Natural Science Foundation of China (NSFC)[11988102] ; NSFC project[11972038] ; NSFC project[12002345] ; National Key Project[GJXM92579] ; Strategic Priority Research Program[XDB22040104]
项目资助者	National Natural Science Foundation of China (NSFC) ; NSFC project ; National Key Project ; Strategic Priority Research Program
论文分区	二类
力学所作者排名	1
RpAuthor	Yang, Zixuan
引用统计	被引频次：5[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://dspace.imech.ac.cn/handle/311007/89316
专题	非线性力学国家重点实验室
作者单位	1.Chinese Acad Sci, Inst Mech, State Key Lab Nonlinear Mech, Beijing 100190, Peoples R China; 2.Univ Chinese Acad Sci, Sch Engn Sci, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Zhou ZD,Li BL,Yang XL,et al. A robust super-resolution reconstruction model of turbulent flow data based on deep learning[J]. COMPUTERS & FLUIDS,2022,239:15.
APA	周志登,李秉霖,杨晓雷,&杨子轩.(2022).A robust super-resolution reconstruction model of turbulent flow data based on deep learning.COMPUTERS & FLUIDS,239,15.
MLA	周志登,et al."A robust super-resolution reconstruction model of turbulent flow data based on deep learning".COMPUTERS & FLUIDS 239(2022):15.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Jp2022FA366_2022_A r（4959KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载