Simulation and Modelling of Turbulent Trailing-Edge Flow

Computations of turbulent trailing-edge flow have been carried out at a Reynolds number of 1000 (based on the free-stream quantities and the trailing-edge thickness) using an unsteady 3D Reynolds-Averaged Navier–Stokes (URANS) code, in which two-equation (k–ε) turbulence models with various low-Re near wall treatments were implemented. Results from a direct numerical simulation (DNS) of the same flow are available for comparison and assessment of the turbulence models used in the URANS code. Two-dimensional URANS calculations are carried out with turbulence mean properties from the DNS used at the inlet; the inflow boundary-layer thickness is 6.42 times the trailing-edge thickness, close to typical turbine blade flow applications. Many of the key flow features observed in DNS are also predicted by the modelling; the flow oscillates in a similar way to that found in bluff-body flow with a von Kármán vortex street produced downstream. The recirculation bubble predicted by unsteady RANS has a similar shape to DNS, but with a length only half that of the DNS. It is found that the unsteadiness plays an important role in the near wake, comparable to the modelled turbulence, but that far downstream the modelled turbulence dominates. A spectral analysis applied to the force coefficient in the wall normal direction shows that a Strouhal number based on the trailing-edge thickness is 0.23, approximately twice that observed in DNS. To assess the modelling approximations, an a priori analysis has been applied using DNS data for the key individual terms in the turbulence model equations. A possible refinement to account for pressure transport is discussed. This revised version was published online in July 2006 with corrections to the Cover Date.     

基于流动仿真大数据应对旋涡-湍流的研究进展

文章以流体科学进入二十一世纪后，在大规模超算、云存储、数据通信和人工智能为支撑的大数据时代背景下，结合目前在复旦大学航空航天系所构建的热流体湍流直接数值仿真数据库，以及复旦大学团队近期与美国德州大学刘超群教授、上海理工大学蔡小舒教授以及国内水动力学杂志编辑部所合作开展的第三代涡识别技术研究，初步概念性地展示旋涡和湍流，特别是针对有工程实际背景和直接应用价值的壁湍流，在这两个流体力学关键基础议题上的最新认知，和基于大数据深度学习的相关湍流工程模拟实践成果．这些成果包括：(1) 基于第三代涡识别技术的尾迹湍流中的涡运动学和动力学探索；(2) 流-热统一完整的类-1、类-2湍流边界层壁面律构建；(3) 基于第三代涡识别量对Kolmogorov 幂次律的再认知；(4) 基于DNS统计数据和神经网络深度学习构建新型湍流封闭模型及RANS计算实践．通过这些成果展示，论证解决这两个基础流体科学议题的技术路径，进而促进流体及相关学科研究在现代大数据背景下取得实质性进展和突破，并惠及现代流体、气动、水利、动力和化工等工程领域．     

走近湍流

简要回顾了20世纪湍流研究的进展,特别是近30多年湍流研究的新成就。对于湍流研究的若干热点问题提出了一些看法。     

Modeling and Simulation of Effects of Turbulence on Vaporization, Mixing and Combustion of Liquid-Fuel Sprays

The objective of this work is twofold. Firstly, the effects of turbulence intensity variations on the turbulent droplet dispersion, vaporization and mixing for non-reacting sprays (with and without swirl) are pointed out. Secondly, the effects of the coupling of the turbulence modulation with external parameters, such as swirl intensity, on turbulent spray combustion are analyzed in configurations of engineering importance. This is achieved by using advanced models for turbulence, evaporation and turbulence modulation implemented into FASTEST-LAG3D-codes: (1) To highlight the influence of turbulence modulation on some spray properties, a thermodynamically consistent modulation model has been considered besides the standard assumption and the well known Crowe's model. For turbulent droplet dispersion, we rely on the Markov-sequence formulation. (2) In order to characterize phase transition processes ongoing on droplets surfaces, a non-equilibrium evaporation model shows better agreement with experiments in comparison with the quasi-equilibrium-based evaporation models often used. (3) The results of turbulence intensity variations reveal the existence of a limited range out of which the increase or decrease of the turbulence intensity affects no more the efficiency of the heat and mass transfer. A derived characteristic number, a vaporization Damkhöler number, possesses a critical value which separates two different behavior regimes with respect to the turbulence/droplet vaporization interactions. (4) Under reacting conditions, it is shown how the evaporation characteristics, mixing rate and combustion process are strongly influenced by swirl intensity and turbulence modulation. In particular, the turbulence modulation modifies the evaporation rate, which in turn influences the mixing and the species concentration distribution. In the case under investigation, it is demonstrated that this effect cannot be neglected for low swirl intensities (Sw.Nu. ≤ 1) in the region far from the nozzle, and close to the nozzle for high swirl number intensities. In providing these particular characteristics, a reliable control of the mixing of gaseous fuel and air in evaporating and reacting sprays, and a possible optimization of the mixing process can tentatively be achieved.     

RESEARCH ON IMPROVEMENTS OF LRN TURBULENCE MODEL BASED ON FLOW AROUND AUTOMOBILE BODY 1)

本文将汽车绕流模块化为各典型局部流动,通过常用湍流模型对各典型局部流动进行数值模拟,结果验证了湍流模型对转捩的捕捉能力是准确模拟汽车绕流的关键. 在分析汽车绕流分离及转捩机理的基础上,优化了稳态和瞬态求解方法,改进了湍流模型对转捩的预测能力,进而提高了湍流模型在汽车流场模拟上的精度. 针对汽车绕流的稳态问题,将流线曲率因子及 响应阈值引入 LRN $k$-$\varepsilon $ (low Reynolds number $k$-$\varepsilon $) 模型,获得了一种能够更准确预 测转捩的改进低雷诺数湍流模型 (modified LRN $k$-$\varepsilon $),改善了原模型对湍流耗散率的过强依赖性及全应力发展预测不足等问题;针对汽车绕流瞬态求解,通过分析 RANS/LES 混合湍流模型的构造思想及特点,引入约束大涡模拟方法,结合本文提出的改进的 LRN $k$-$\varepsilon $ 湍流模型,提出了一种能准确捕捉转捩现象 的转捩 LRN CLES 模型. 分别将改进的模型用于某实车外流场和风振噪声仿真中,通过 Ansys Fluent 求解器计算,并将计算结果与常用湍流模型的仿真结果、HD-2 风洞试验结果和实车道路实验结果进行对比,表明改进后的湍流模型能够更准确模拟复杂实车的稳态和瞬态特性,为汽车气动特性的研究提供了可靠理论依据及有效数值解决方法.     

Linear and nonlinear inversion schemes to retrieve collision kernel values from droplet size distribution change

This study presents an attempt to retrieve collision kernel values from changes in the droplet size distribution due to collision growth. Original linear and nonlinear inversion schemes are presented, which use the simple a priori assumption that the total collision rate is given by the sum of the gravitational and turbulent contributions. Our schemes directly handle binned (discretized) size distributions and, therefore, do not require any assumptions on distribution functional forms, such as the self-similarity assumption. To validate the schemes, three-dimensional direct numerical simulation (DNS) of colliding droplets in steady isotropic turbulence is performed. In the DNS, air turbulence is calculated using a pseudo-spectral method, while droplet motions are tracked by the Lagrangian method. Comparison between the retrieved collision kernels and the collision kernels obtained directly from the DNS show that for low Reynolds number flows both the linear and nonlinear inversion schemes give good accuracy. However, for higher Reynolds number flows the linear inversion scheme gives significantly larger retrieval errors, while the errors for the nonlinear scheme remain small.     

On the performance of sudden‐expansion pipe without/with cross‐flow injection: experimental and numerical studies

The paper explores the possibilities that different turbulence closures offer, for in‐depth analysis of a complex flow. The case under investigation is steady, turbulent flow in a pipe with sudden expansion without/with normal‐to‐wall injection through jets. This is a typical geometry where generation of turbulence energy takes place, due to sudden change in boundary conditions. This study is aimed at investigating the capability of a developed computational program by the present authors with three different turbulence models to calculate the mean flow variables. Three two‐equation models are implemented, namely the standard linear k ? ε model, the low Reynolds number k ? ε model and the cubic nonlinear eddy viscosity (NLEV) k ? ε model. The performance of the chosen turbulence models is investigated with regard to the available data in the literature including velocity profiles, turbulent kinetic energy and reattachment position in a pipe expansion. In order to further assess the reliability of the turbulence models, an experimental program was conducted by the present authors also at the fluid mechanics laboratory of Menoufiya University. Preliminary measurements, including the surface pressure along the two walls of the expansion pipe and the pressure drop without and with the presence of different arrangements of wall jets produced by symmetrical or asymmetrical fluid cross‐flow injection, are introduced. The results of the present studies demonstrate the superiority of the cubic NLEV k ? ε model in predicting the flow characteristics over the entire domain. The simple low Reynolds number k ? ε model also gives good prediction, especially when the reattachment point is concerned. The evaluation of the reattachment point and the pressure‐loss coefficient is numerically addressed in the paper using the cubic NLEV k ? ε model. The results show that the injection location can control the performance of the pipe‐expansion system. It is concluded that the introduction of flow injection can increase the energy loss in the pipe expansion. The near‐field turbulence structure is also considered in the present study and it is noticed that the turbulence level is strongly affected by the cross‐flow injection and the jet location. Copyright © 2011 John Wiley & Sons, Ltd.     

壁湍流相干结构和减阻控制机理

剪切湍流中相干结构的发现是上世纪湍流研究的重大进展之一,这些大尺度的相干运动在湍流的动力学过程中起重要作用,也为湍流的控制指出了新的方向.壁湍流高摩擦阻力的产生与近壁区流动结构密切相关,基于近壁区湍流动力学过程的减阻控制方案可以有效降低湍流的摩擦阻力,但是随着雷诺数的升高, 这些控制方案的有效性逐渐降低.近年来研究发现, 在高雷诺数情况下外区存在大尺度的相干运动,这种大尺度运动对近壁区湍流和壁面摩擦阻力的产生有重要影响,为高雷诺数湍流减阻控制策略的设计提出了新的挑战.该文将对壁湍流相干结构的研究历史加以简单的回顾,重点介绍近壁区相干结构及其控制机理、近年来高雷诺数外区大尺度运动的研究进展,在此基础上提出高雷诺数减阻控制研究的关键科学问题.      

Decaying homogeneous turbulence under strong density stratification at low-Prandtl number

The characteristics of decaying homogeneous turbulence under strong density stratification have been studied using direct numerical simulations. While our previous study dealt with rotating stratified turbulence, here we investigate the detailed flow structure of stratified turbulence without rotation especially at low-Prandtl number. By assuming a low-Prandtl-number fluid, e.g. liquid sodium: Pr ≈ 0.01, gallium: Pr ≈ 0.025, internal gravity waves are markedly attenuated due to the large thermal conductivity, and turbulence soon reaches a two-component state, where vertical energy, coupled with potential energy, significantly decays, and becomes negligible as observed experimentally (Praud et al. in J Fluid Mech 522:1–33, 2005). In the horizontal plane, there appear large-scale vortices with vertical vorticity, and those with the same sign of vorticity increase their horizontal length scale by merging with each other. In the vertical plane, highly sheared regions represented by horizontal vorticity also tend to horizontally increase their length scale and become layered structures by the combined effects of vortex coalescence and energy cascade into higher vertical wavenumbers.      

横向紊动射流的数值与实验研究进展

横向紊动射流作为流体运动的一种重要类型,广泛存在于如: 燃气轮机气膜冷却、锅炉燃烧室等的燃烧控制, V/STOL(垂直或短距离起落)飞机、废气排放的控制等工程实际应用中.由于射流的存在,增加了流场的复杂性,流场中同时存在射流剪切层涡、马蹄形涡系、反向旋涡对和尾迹涡等4种涡系结构,这对流体力学理论研究具有重要意义.长期以来,研究人员从理论分析、实验测量和数值模拟方面对横向紊动射流进行了大量的研究工作,目前已经认识了流场中的许多流动特性和流动机理.从数值模拟和实验研究两个方面,比较并分析了国内外横向紊动射流研究的现状和研究结果,评述了不同湍流模型以及不同的实验测量方法对横向紊动射流的预测能力,讨论了存在的问题并对该领域的研究方向进行了展望.      

Numerical simulation of turbulence suppression: Comparisons of the performance of four k-? turbulence models

The standard k-ε model and three low-Reynolds number k-ε models were used to simulate pipe flow with a ring device installed in the near-wall region. Both developing and fully developed turbulent pipe flows have been investigated. Turbulence suppression for fully developed pipe flows revealed by hot-wire measurements has been predicted with all three low-Reynolds number models, and turbulence enhancement has been predicted by the standard k-ε model. All three low-Reynolds number models have predicted similar distributions of velocities, turbulence kinetic energy, and dissipation rate. For developing pipe flows, the region of turbulence suppression predicted by the three low-Reynolds number models is much more extensive (up to 30 pipe diameters downstream of the device) than for full developed flow; whereas the standard k-ε model has only predicted turbulence enhancement.     

On the evolution of decaying isotropic turbulence

A direct numerical simulation is performed on 256³ grids for decaying isotropic turbulence. The total kinematic energy, Taylor micro-scale, Taylor micro-scale Reynolds number and the velocity derivative skewness are calculated. The snapshots of energy spectra and energy transfer spectra are plotted. These measurements verify the DIA predictions: decaying isotropic turbulence has the energy propagation and occupies the final decay periods. The skewness remains to some level with small variation even in the final decay period.     

Band-pass filtered velocity statistics in decaying turbulent box

For homogeneous isotropic turbulence study, the acquisition of band-pass filtered velocity increments (FVI) in a non-forced turbulent box is still a challenge both experimentally and numerically. Turbulence and associated physical processes, at a given instant, are permanently contaminated by a forcing process which can seldom be universal. The situation tends to be the origin of intermittency and the non-Gaussian probability density distribution for acceleration and velocity gradients. To reveal implied mechanism, grid turbulence is adapted to observe non-perturbed homogeneous isotropic turbulence. The velocity increments (VI) can be obtained following Comte-Bellot and Corrsin (GCBC) by means of two point-two time shifted velocity measurements. It is difficult to obtain decaying turbulence (DT) at large turbulent Reynolds number without pollution coming from walls. Nevertheless it is also significant to investigate DT in low Reynolds number regimes to determine non-polluted tendencies. The similarity of DT between particle image velocimetry (PIV) and hot wire anemometry measurements by GCBC are presented. Here we focus our tendency on VI and FVI probability density function (PDF) shapes in this letter. In conclusion, the tendency to Gaussian shape in inertial zone wavenumbers, demonstrates that there will be no intermittency if turbulent cascade is not perturbed.     

Pulsed-wire anemometry

Pulsed-wire anemometry was first developed as a tool for making velocity and turbulence measurements in the late 1960s. It has been continuously refined and its potential exploited by a number of groups who have obtained reliable data in situations where the use of other techniques would have been extremely difficult. The technique is particularly useful in flows of high turbulence intensity and has therefore been used to greatest effect in separated flows. Although its range of applicability is much more restricted than that of laser-Doppler anemometry, it is an order of magnitude cheaper, and it does not require seeding of the flow, with all the attendant uncertainties.

First the physical basis of pulsed-wire anemometry is briefly described, and the major sources of experimental error are outlined for cases in which the technique is used both remote from and close to walls. Progress in the design of probes, which have been improved in a number of ways since the early days to reduce errors, is also outlined. The author then reviews the kinds of measurements that have been successfully made and the consequent improvements in the understanding of the physics of complex flows. Examples are drawn from a wide range of work, including some unusual applications like measurements in very low velocity gas mixtures. The paper closes with a summary of the limitations of the technique and an overall assessment of the likely potential for its use in future turbulence research.     

Hyperviscosity and Vorticity-Based Models for Subgrid Scale Modeling

Large-eddy simulations corresponding to the decaying isotropic turbulence experiment of Comte-Bellot and Corrsin are performed, using a pseudo-spectral code that incorporates four models: viscosity and hyperviscosity types, each implemented for both the subgrid scale stress tensor and the subgrid scale force. Two 1/T scalings are also considered for the viscosity amplitude. The dynamic procedure is extended to the four models and is tested. Results are obtained with and without this procedure and for both scalings. The main conclusions are: (a) the two viscosity models perform equally well; (b) the Kolmogorov scaling performs as well as the Smagorinsky scaling, yet it is computationally more efficient; (c) in the dynamic procedure, there is a fairly wide range of test to grid filter ratios which produces results insensitive to this ratio; and (d) the hyperviscosity models lead to energy decay curves that follow the experimental data as well as the usual viscosity models.     

Turbulent flow around a wall-mounted cube: A direct numerical simulation

An immersed-boundary method was employed to perform a direct numerical simulation (DNS) of flow around a wall-mounted cube in a fully developed turbulent channel for a Reynolds number Re = 5610, based on the bulk velocity and the channel height. Instantaneous results of the DNS of a plain channel flow were used as a fully developed inflow condition for the main channel. The results confirm the unsteadiness of the considered flow caused by the unstable interaction of a horseshoe vortex formed in front of the cube and on both its sides with an arch-type vortex behind the cube. The time-averaged data of the turbulence mean-square intensities, Reynolds shear stresses, kinetic energy and dissipation rate are presented. The negative turbulence production is predicted in the region in front of the cube where the main horseshoe vortex originates.     

Numerical study of physiological turbulent flows through series arterial stenoses

A numerical investigation on the characteristics of transitional turbulent flow over series bell‐shape stenoses for a physiological pulsatile flow is presented in the present study. The flow behaviours for the physiological pulsatile flow are studied by considering the effects of the Reynolds number, Womersley number, constriction ratio and spacing ratio of the stenoses on the pulsatile turbulent flow fields. Especially, the mutual influences between the double stenoses under different flow conditions are considered. The numerical results show that the variation of these flow parameters puts significant impacts on the flow developments in the arteries with series stenoses. The double stenoses lead to the higher peak turbulence disturbance and the greater area with comparatively high turbulence intensity distal to the stenoses in comparison with the single stenosis. The analysis shows that for the physiological pulsatile flow, the downstream stenosis usually does not have perceptible influences on the upstream flow fields. Copyright © 2004 John Wiley & Sons, Ltd.     

Large eddy simulation of free surface shallow‐water flow

Shallow‐water flow with free surface frequently occurs in ambient water bodies, in which the horizontal scale of motion is generally two orders of magnitude greater than the water depth. To accurately predict this flow phenomenon in more detail, a three‐dimensional numerical model incorporating the method of large eddy simulation (LES) has been developed and assessed. The governing equations are split into three parts in the finite difference solution: advection, dispersion and propagation. The advection part is solved by the QUICKEST scheme. The dispersion part is solved by the central difference method and the propagation part is solved implicitly using the Gauss–Seidel iteration method. The model has been applied to free surface channel flow for which ample experimental data are available for verification. The inflow boundary condition for turbulence is generated by a spectral line processor. The computed results compare favourably with the experimental data and those results obtained by using a periodic boundary condition. The performance of the model is also assessed for the case in which anisotropic grids and filters with horizontal grid size of the order of the water depth are used for computational efficiency. The coarse horizontal grid was found to cause a significant reduction in the large‐scale turbulent motion generated by the bottom turbulence, and the turbulent motion is predominately described by the sub‐grid scale (SGS) terms. The use of the Smagorinsky model for SGS turbulence in this situation is found inappropriate. A parabolic mixing length model, which accounts for the filtered turbulence, is then proposed. The new model can reproduce more accurately the flow quantities. Copyright © 2000 John Wiley & Sons, Ltd.     

Assessment of the modulated gradient model in decaying isotropic turbulence

A recently introduced nonlinear model undergoes evaluations based on two isotropic turbulent cases: a University of Wiscosion-Madison case at a moderate Reynolds number and a Johns Hopkins University case at a high Reynolds number. The model uses an estimation of the subgrid-scale (SGS) kinetic energy to model the magnitude of the SGS stress tensor, and uses the normalized velocity gradient tensor to model the structure of the SGS stress tensor. Testing is performed for the first case through a comparison between direct numerical simulation (DNS) results and large eddy simulation (LES) results regarding resolved kinetic energy and energy spectrum. In the second case, we examine the resolved kinetic energy, the energy spectrum, as well as other key statistics including the probability density functions of velocities and velocity gradients, the skewness factors, and the flatness factors. Simulations using the model are numerically stable, and results are satisfactorily compared with DNS results and consistent with statistical theories of turbulence.