Large Eddy Simulation of Stratified Turbulent Channel Flow with a Dynamic Subgrid Scale Model

In this paper, large eddy simulation coupled with a dynamic subgrid scale (SGS) model is used to study turbulent channel flows with heat transfer. Some fundamental flow behaviors are discussed. Based on our calculated results, the dynamic SGS model can reasonably predict some main characteristics of stratified turbulent channel flows. Our results are also in good agreement with theoretical analyses and previous calculated results.

Large Eddy Simulations of Mixing Time in a Stirred Tank

Large eddy simulations (LES) of mixing process in a stirred tank of 0.476m diameter with a 3-narrow blade hydrofoil CBY impeller were reported. The turbulent flow field and mixing time were calculated using LES with Sma-gorinsky-Lilly subgrid scale model. The impeller rotation was modeled using the sliding mesh technique. Better agree-ment of power demand and mixing time was obtained between the experimental and the LES prediction than that by the traditional Reynolds-averaged Navier-Stokes (RANS) approach. The curve of tracer response predicted by LES was in good agreement with the experimental. The results show that LES is a reliable tool to investigate the unsteady and quasi-periodic behavior of the turbulent flow in stirred tanks.

LES prediction of space-time correlations in turbulent shear flows

We compare the space-time correlations calculated from direct numerical simulation （DNS） and large-eddy simulation （LES） of turbulent channel flows. It is found from the comparisons that the LES with an eddy-viscosity subgrid scale （SGS） model over-predicts the space-time corre- lations than the DNS. The overpredictions are further quantified by the integral scales of directional correlations and convection velocities. A physical argument for the overpre- diction is provided that the eddy-viscosity SGS model alone does not includes the backscatter effects although it correctly represents the energy dissipations of SGS motions. This argument is confirmed by the recently developed elliptic model for space-time correlations in turbulent shear flows. It suggests that enstrophy is crucial to the LES prediction of spacetime correlations. The random forcing models and stochastic SGS models are proposed to overcome the overpredictions on space-time correlations.

LES Investigation of Drag-Reducing Mechanism of Turbulent Channel Flow with Surfactant Additives

In this work,the drag-reducing mechanism of high-Reynoldsnumber turbulent channel flow with surfactant additives is investigated by using large eddy simulation(LES)method.An N-parallel finitely extensible nonlinear elastic model with Peterlin’s approximation(FENE-P)is used to describe the rheological behaviors of non-Newtonian fluid with surfactant.To close the filtered LES equations,a hybrid subgrid scale(SGS)model coupling the spatial filter and temporal filter is applied to compute the subgrid stress and other subfilter terms.The finite difference method and projection algorithm are adopted to solve the LES governing equations.To validate the correctness of our LES method and in-house code,the particle image velocimetry(PIV)experiment is carried out and representative measured results are compared with LES results in detail.Then the flow characteristics and drag-reducing mechanism of turbulent channel flow with surfactant are investigated from the perspective of drag reduction rate,mean velocity,fluctuation of deformation rate,shear stress,transport and dissipation of turbulent kinetic energy,and turbulent coherent structures.This research can shed a light on the application of turbulent drag reduction technique in district heating,petroleum transport,etc.

A NEW STABILIZED SUBGRID EDDY VISCOSITY METHOD BASED ON PRESSURE PROJECTION AND EXTRAPOLATED TRAPEZOIDAL RULE FOR THE TRANSIENT NAVIER-STOKES EQUATIONS

We consider a new subgrid eddy viscosity method based on pressure projection and extrapolated trapezoidal rule for the transient Navier-Stokes equations by using lowest equal-order pair of finite elements. The scheme stabilizes convection dominated problems and ameliorates the restrictive inf-sup compatibility stability. It has some attractive fea- tures including parameter free for the pressure stabilized term and calculations required for higher order derivatives. Moreover, it requires only the solutions of the linear system arising from an Oseen problem per time step and has second order temporal accuracy. The method achieves optimal accuracy with respect to solution regularity.

A Liutex-based subgrid stress model for large-eddy simulation

The concept of vortex is crucial in both understanding and modeling of turbulence.For large eddy simulation(LES),the effect of small-scale eddies onto the large scales or the resolved flow field is modeled by subgrid stress models.Even though the rotating motions of fluids,i.e.,vortices or eddies are central in developing turbulent models,vortex identification methods are seldom used in these models.In this study,we develop a new subgrid model based on the Liutex vector,a new quantity introduced to decompose fluid motions into rigid rotation,pure shear and stretching,and thus identify vortices.The new model is then applied in a decaying homogeneous isotropic turbulence(DHIT)and a turbulent channel flow at Reynolds number Reτ=180.It is shown that the new model can predict accurate energy spectra compared with experiments in DHIT and give a well-matched velocity profile in turbulent channel flow without changing the form of the model.Future directions include improvement of the Liutex based model,for example developing anisotropic subgrid models,and its applications in various turbulent flows.

A LATTICE BOLTZMANN SUBGRID MODEL FOR LID-DRIVEN CAVITY FLOW

In recent years, the Lattice Boltzmann Method (LBM) has developed into an alternative and promising numerical scheme for simulating fluid flows and modeling physics in fluids. In order to propose LBM for high Reynolds number fluid flow applications, a subgrid turbulence model for LBM was introduced based on standard Smagorinsky subgrid model and Lattice Bhatnagar-Gross-Krook (LBGK) model. The subgrid LBGK model was subsequently used to simulate the two-dimensional driven cavity flow at high Reynolds numbers. The simulation results including distribution of stream lines, dimensionless velocities distribution, values of stream function, as well as location of vertex center, were compared with benchmark solutions, with satisfactory agreements.

AN INVESTIGATION OF TURBULENT HEAT TRANSFER IN CHANNEL FLOWS BY LARGE EDDY SIMULATION

Large Eddy Simulation (LES) of fully developed turbulent channel flow with heat transfer was performed to investigate the effects of the Reynolds number on the turbulence behavior. In the present study, the bottom wall of the channel was cooled and the top wall was heated. The Reynolds numbers, based on the central mean-velocity and the half-width of the channel, were chosen as 4000, 6000, 10 4 and 2×10 4, and the Prandtl number as 1.0. To validate our calculations, the present results were compared with available data obtained by Direct Numerical Simulation (DNS), which proves to be in good agreement with each other. To reveal the effects of the Reynolds number, some typical quantities, including the velocity fluctuations, temperature fluctuation, heat fluxes and turbulent Prandtl number, were studied.

Macro and micro issues in turbulent mixing

Numerical prediction of turbulent mixing can be divided into two subproblems: to predict the geometrical extent of a mixing region and to predict the mixing properties on an atomic or molecular scale, within the mixing region. The former goal suffices for some purposes, while important problems of chemical reactions(e.g. flames) and nuclear reactions depend critically on the second goal in addition to the first one. Here we review recent progress in establishing a conceptual reformulation of convergence, and we illustrate these concepts with a review of recent numerical studies addressing turbulence and mixing in the high Reynolds number limit. We review significant progress on the first goal, regarding the mixing region, and initial progress on the second goal, regarding atomic level mixing properties. New results concerning non-uniqueness of the infinite Reynolds number solutions and other consequences of a renormalization group point of view, to be published in detail elsewhere, are summarized here.The notion of stochastic convergence(of probability measures and probability distribution functions) replaces traditional pointwise convergence. The primary benefit of this idea is its increased stability relative to the statistical "noise" which characterizes turbulent flow. Our results also show that this modification of convergence, with sufficient mesh refinement, may not be needed. However, in practice, mesh refinement is seldom sufficient and the stochastic convergence concepts have a role.Related to this circle of ideas is the observation that turbulent mixing, in the limit of high Reynolds number, appears to be non-unique. Not only have multiple solutions been observed(and published) for identical problems, but simple physics based arguments and more refined arguments based on the renormalization group come to the same conclusion.Because of the non-uniqueness inherent in numerical models of high Reynolds number turbulence and mixing, we also include here numerical examples of validation. The algorithm we use here has two essential components. We depend on Front Tracking to allow accurate resolution of flows with sharp interfaces or steep gradients(concentration or thermal), as are common in turbulent mixing problems. The higher order and enhanced algorithms for interface tracking, both those already developed, and those proposed here, allow a high resolution and uniquely accurate description of sample mixing problems. Additionally, we depend on the use of dynamic subgrid scale models to set otherwise missing values for turbulent transport coefficients, a step that breaks the non-uniqueness.