Ensemble Kalman method for learning turbulence models from indirect observation data

doi:10.1017/jfm.2022.744

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

	Ensemble Kalman method for learning turbulence models from indirect observation data
	Zhang XL(张鑫磊); Xiao, Heng 3; Luo, Xiaodong 2; He GW(何国威)
刊名	JOURNAL OF FLUID MECHANICS
	2022-09
卷号	949
关键词	turbulence modelling machine learning
ISSN号	0022-1120
DOI	10.1017/jfm.2022.744
英文摘要	In this work, we propose using an ensemble Kalman method to learn a nonlinear eddy viscosity model, represented as a tensor basis neural network, from velocity data. Data-driven turbulence models have emerged as a promising alternative to traditional models for providing closure mapping from the mean velocities to Reynolds stresses. Most data-driven models in this category need full-field Reynolds stress data for training, which not only places stringent demand on the data generation but also makes the trained model ill-conditioned and lacks robustness. This difficulty can be alleviated by incorporating the Reynolds-averaged Navier-Stokes (RANS) solver in the training process. However, this would necessitate developing adjoint solvers of the RANS model, which requires extra effort in code development and maintenance. Given this difficulty, we present an ensemble Kalman method with an adaptive step size to train a neural-network-based turbulence model by using indirect observation data. To our knowledge, this is the first such attempt in turbulence modelling. The ensemble method is first verified on the flow in a square duct, where it correctly learns the underlying turbulence models from velocity data. Then the generalizability of the learned model is evaluated on a family of separated flows over periodic hills. It is demonstrated that the turbulence model learned in one flow can predict flows in similar configurations with varying slopes.
学科主题	Mechanics ; Physics, Fluids & Plasmas
分类号	一类/力学重要期刊
语种	英语
WOS记录号	WOS:000861459400001
资助机构	NSFC Basic Science Center Program for `Multiscale Problems in Nonlinear Mechanics' [11988102] ; National Natural Science Foundation of China [12102435] ; China Postdoctoral Science Foundation [2021M690154] ; National Centre for Sustainable Subsurface Utilization of the Norwegian Continental Shelf, Norway [NCS2030]
其他责任者	He, GW (corresponding author), Chinese Acad Sci, Inst Mech, State Key Lab Nonlinear Mech, Beijing 100049, Peoples R China. ; He, GW (corresponding author), Univ Chinese Acad Sci, Sch Engn Sci, Beijing 100049, Peoples R China. ; Xiao, H (corresponding author), Virginia Tech, Kevin T Crofton Dept Aerosp & Ocean Engn, Blacksburg, VA 24060 USA.
内容类型	期刊论文
源URL	[http://dspace.imech.ac.cn/handle/311007/90181]
专题	力学研究所_非线性力学国家重点实验室
作者单位	1.Norwegian Res Ctr NORCE, Nygardsgaten 112, N-5008 Bergen, Norway 2.Virginia Tech, Kevin T Crofton Dept Aerosp & Ocean Engn, Blacksburg, VA 24060 USA 3.Univ Chinese Acad Sci, Sch Engn Sci, Beijing 100049, Peoples R China 4.Chinese Acad Sci, Inst Mech, State Key Lab Nonlinear Mech, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Zhang XL,Xiao, Heng,Luo, Xiaodong,et al. Ensemble Kalman method for learning turbulence models from indirect observation data[J]. JOURNAL OF FLUID MECHANICS,2022,949.
APA	张鑫磊,Xiao, Heng,Luo, Xiaodong,&何国威.(2022).Ensemble Kalman method for learning turbulence models from indirect observation data.JOURNAL OF FLUID MECHANICS,949.
MLA	张鑫磊,et al."Ensemble Kalman method for learning turbulence models from indirect observation data".JOURNAL OF FLUID MECHANICS 949(2022).