Data augmented prediction of Reynolds stresses for flows around an axisymmetric body of revolution

doi:10.1016/j.oceaneng.2024.116717

CORC > 力学研究所 > 中国科学院力学研究所 > 非线性力学国家重点实验室

	Data augmented prediction of Reynolds stresses for flows around an axisymmetric body of revolution
	Liu, Yi ; Wang, Shizhao ; Zhang, Xin-Lei 2; He, Guowei 2
刊名	OCEAN ENGINEERING
	2024-03-15
卷号	296 页码:14
关键词	Ensemble Kalman method Machine learning Turbulence modeling SUBOFF model
ISSN号	0029-8018
DOI	10.1016/j.oceaneng.2024.116717
通讯作者	Zhang, Xin-Lei(zhangxinlei@imech.ac.cn) ; He, Guowei(hgw@lnm.imech.ac.cn)
英文摘要	This work presents a data -driven approach to improve the predictive accuracy of the Reynolds stresses for flows over an axisymmetric body of revolution. Sparse experimental data of velocity and Reynolds normal stress is used to learn a nonlinear eddy viscosity model represented by neural networks with the ensemble Kalman method. It was recently proposed in Zhang et al., (2023) to enhance the predictive capability of the model by introducing a neural network -based model correction in the turbulent kinetic energy transport equation. Here this newly developed technique is applied for the first time to three-dimensional engineering flows in which the complete set of tensor functions are used. The results show that the method can learn a model with good predictive capability in both velocity and turbulence kinetic energy. The predictive improvement is achieved by suppressing turbulence production in the boundary layer and wake -shear layer with the added correction term in the turbulent kinetic energy transport equation. Moreover, the learned model can improve the prediction of Reynolds normal stresses by capturing the Reynolds stress anisotropy with nonlinear tensor bases. The necessity of combining the nonlinear tensor functions and the correction in turbulence transport equations is highlighted for accurate prediction of the Reynolds stress.
资助项目	NSFC Basic Science Center Program[11988102] ; CAS Project for Young Scientists in Basic Research[YSBR-087] ; National Natural Science Foundation of China[12102439] ; National Natural Science Foundation of China[12102435] ; China Postdoctoral Science Foundation[2021M703290] ; China Postdoctoral Science Foundation[2021M690154]
WOS关键词	LARGE-EDDY SIMULATION ; TURBULENCE MODELS ; NOISE
WOS研究方向	Engineering ; Oceanography
语种	英语
WOS记录号	WOS:001207409800001
资助机构	NSFC Basic Science Center Program ; CAS Project for Young Scientists in Basic Research ; National Natural Science Foundation of China ; China Postdoctoral Science Foundation
内容类型	期刊论文
源URL	[http://dspace.imech.ac.cn/handle/311007/94987]
专题	力学研究所_非线性力学国家重点实验室
通讯作者	Zhang, Xin-Lei; He, Guowei
作者单位	1.Univ Chinese Acad Sci, Sch Engn Sci, Beijing 100049, Peoples R China 2.Chinese Acad Sci, LNM, Inst Mech, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Liu, Yi,Wang, Shizhao,Zhang, Xin-Lei,et al. Data augmented prediction of Reynolds stresses for flows around an axisymmetric body of revolution[J]. OCEAN ENGINEERING,2024,296:14.
APA	Liu, Yi,Wang, Shizhao,Zhang, Xin-Lei,&He, Guowei.(2024).Data augmented prediction of Reynolds stresses for flows around an axisymmetric body of revolution.OCEAN ENGINEERING,296,14.
MLA	Liu, Yi,et al."Data augmented prediction of Reynolds stresses for flows around an axisymmetric body of revolution".OCEAN ENGINEERING 296(2024):14.