Adjoint-based computation of nonlocal eddy viscosity in turbulent channel flow

被引:0
|
作者
Liu, Jessie [1 ]
Bryngelson, Spencer H. [2 ,3 ]
Zaki, Tamer A. [4 ]
Mani, Ali [1 ]
机构
[1] Stanford Univ, Dept Mech Engn, Stanford, CA 94305 USA
[2] Georgia Inst Technol, Sch Computat Sci & Engn, Atlanta, GA 30332 USA
[3] Georgia Inst Technol, Daniel Guggenheim Sch Aerosp Engn, Atlanta, GA 30332 USA
[4] Johns Hopkins Univ, Dept Mech Engn, Baltimore, MD 21218 USA
来源
PHYSICAL REVIEW FLUIDS | 2024年 / 9卷 / 09期
关键词
SCALAR TRANSPORT; MODELS;
D O I
10.1103/PhysRevFluids.9.094606
中图分类号
O35 [流体力学]; O53 [等离子体物理学];
学科分类号
070204 ; 080103 ; 080704 ;
摘要
Reynolds-averaged Navier-Stokes (RANS) closure operators are generally nonlocal and anisotropic for many flows, including wall-bounded turbulence. The macroscopic forcing method (MFM) and Green's function-based approaches examine these effects using forced direct numerical simulations. For these approaches, the number of simulations required to compute the nonlocal and anisotropic eddy viscosity is equal to the number of degrees of freedom in the averaged space. We reduce the computational cost by introducing an adjoint-based formulation of MFM to obtain the nonlocal and anisotropic eddy viscosity at any Reynolds stress location using a single simulation. We then quantify the streamwise and wall-normal nonlocal eddy viscosity at near-wall locations in turbulent channel flow at Re-tau = 180. We demonstrate that the upstream nonlocality is not fully described by Lagrangian transport. We also quantify the significant differences in the upstream nonlocality between various components of the eddy viscosity tensor. Our results can be used to guide closure modeling, including the lengthscales and timescales used in Reynolds stress models.
引用
收藏
页数:31
相关论文
共 50 条
  • [21] Large Eddy Simulation of Spanwise Rotating Turbulent Channel Flow with Subgrid-Scale Eddy Viscosity Model Based on Helicity
    Jiang, Zhou
    Yu, Changping
    ADVANCES IN APPLIED MATHEMATICS AND MECHANICS, 2019, 11 (03) : 711 - 722
  • [22] EDDY VISCOSITY MODEL FOR TURBULENT PIPE FLOW
    KLEINSTEIN, G
    AIAA JOURNAL, 1971, 9 (08) : 1626 - +
  • [23] NEW EDDY VISCOSITY MODEL FOR COMPUTATION OF SWIRLING TURBULENT FLOWS
    KIM, KY
    CHUNG, MK
    AIAA JOURNAL, 1987, 25 (07) : 1020 - 1022
  • [24] Adjoint-based error estimation for grid adaptation for large eddy simulation
    Jiang, Yao
    Nadarajah, Siva
    COMPUTERS & FLUIDS, 2022, 236
  • [25] INVERSE ADJOINT ESTIMATION OF EDDY VISCOSITY FOR COASTAL FLOW MODELS
    PANCHANG, VG
    RICHARDSON, JE
    JOURNAL OF HYDRAULIC ENGINEERING-ASCE, 1993, 119 (04): : 506 - 525
  • [26] Complete adjoint-based incremental power flow equations
    Jesus, C. M. S. C.
    Ferreira, L. A. F. M.
    ELECTRIC POWER SYSTEMS RESEARCH, 2009, 79 (07) : 1136 - 1144
  • [27] A COMPARISON OF ADJOINT-BASED OPTIMIZATIONS FOR INDUSTRIAL PIPE FLOW
    Helgason, Eysteinn
    Krajnovic, Sinisa
    ASME FLUIDS ENGINEERING DIVISION SUMMER MEETING - 2014, VOL 1C: SYMPOSIA, 2014,
  • [28] Compressible effects modeling for turbulent cavitating flow in a small venturi channel: An empirical turbulent eddy viscosity correction
    Zhang, Xin-Lei
    Ge, Ming-Ming
    Zhang, Guang-Jian
    Coutier-Delgosha, Olivier
    PHYSICS OF FLUIDS, 2021, 33 (03)
  • [29] Adjoint-based machine learning for active flow control
    Liu, Xuemin
    Macart, Jonathan F.
    PHYSICAL REVIEW FLUIDS, 2024, 9 (01)
  • [30] Adjoint-based sensitivity and ignition threshold mapping in a turbulent mixing layer
    Capecelatro, Jesse
    Bodony, Daniel J.
    Freund, Jonathan B.
    COMBUSTION THEORY AND MODELLING, 2019, 23 (01) : 147 - 179