用间接边界元方法求解不可压缩孤立翼型绕流的一种库塔条件处理方法

本文对不可压二维翼型势流绕流的边界元法求解作了分析,在对基本方程与边界条件进行数值离散化时,将库塔条件代入基本方程。按本方法编制的计算机程序对若干算例进行了验算。结果表明,本文提出的方法是可靠的,该法计算简捷、方便,占用计算机内存少,具有实用意义。     

自由面势流问题的域外奇点边界元法及其数值误差分析

讨论了域外奇点边界元法在自由面势流问题计算中的作用，并以连续及离散Fourier分析对该方法(就m阶面元的一般情况)进行数值误差分析，导出了计及面元阶数、奇点至自由面垂向距离、配置点移动、差分格式等因素影响的数值误差一般表达式。从理论上证明了自由面势流问题计算中采用域外奇点法可改善离散产生的数值色散误差并能结合配置点前移(向上游)等方法以数值满足辐射条件。     

不可压粘性流体浸入边界直接力法及软件实现

浸入边界法通过在N-S方程中施加体积力模拟不可滑移固壁边界及动边界，避免生成复杂贴体网格及动网格，极大地节省了网格建模时间及动网格计算消耗。本文提出一种新型附加体积力简化计算方法，将简化附加体积力以源项形式嵌入动量方程迭代中，通过用户自定义函数对CFD软件FLUENT二次开发，实现了浸入边界法和通用流体力学求解器的耦合计算。通过静止圆柱和动圆柱绕流数值模拟进行了验证，并探讨了插值函数对计算精度的影响。研究表明，通过引入浸入边界模型，能够提高计算效率，并实现结构网格背景下复杂边界和动边界的高效建模。     

圆柱绕流教学中常被忽略的一个问题

<正> 笔者所见不少流体力学教材在讲述无环量圆柱绕流时,大都这样处理:1)推出速度势(?)(或流函数(?))满足的 Laplace 方程 ▽~2(?)=0(或▽~2(?)=0);2)给出若干个基本解;3)将均匀流与偶极子两个基本解叠     

流体力学中一些未解决问题(Ⅰ)

<正> 前言 本文可能收集有100个问题。将众多研究者的贡献集中起来,这是一次试验。 我们的文献中,并未广泛使用提出问题这种方式。在流体力学中,或者一般地,在科学或工程中,典型的论文不包含一个问题。 有些人明显不愿提出他们尚未解决的问题。但一部分人没有问题,因为提出问题并非我     

侧柱与串列双柱绕流之间的干扰

本文给出了关于串列双柱与创柱间流动干扰的实验研究结果。当三个圆柱排成等边三角形并靠得很近时，由于三圆柱间强烈的缝隙流动，大大地改变了绕流其中的串列双圆柱的流态。特别，当三圆柱中心距等于二倍圆柱直径时，在串列双柱的前、后柱之间形成了强烈的偏斜的缝隙流，出现了独特的压力分布以及要比单柱高出三倍以上的旋涡脱落频率。     

基于滑移速度壁模型的复杂边界湍流大涡模拟

采用滑移速度壁模型实现了浸入边界方法与壁模型相结合的大涡模拟.本文首先分别采用平衡层模型和非平衡壁模型对周期山状流进行数值模拟,以考查在壁模型中考虑切向压力梯度的作用.数值结果表明,流场的压力对本文所采用的壁模型形式并不敏感,但是考虑切向压力梯度可以显著改进壁面摩擦力的计算结果,并且能够准确的预测强压力梯度区以及分离区内的流动平均统计特性.不考虑压力梯度效应的平衡层模型显著低估了壁面摩擦力的分布,同时无法准确预测分离区内的平均速度剖面.非平衡模型的修正项正比于切向压力梯度和壁面法向距离,因此在强压力梯度区或者网格较粗时,计算得到的平均压力和摩擦力分布以及流动的低阶统计量均与参考的实验和计算结果吻合.在此基础上,通过回转体绕流的大涡模拟考查了该方法用于模拟高雷诺数壁湍流的适用性,非平衡壁模型可以准确地捕捉流动的物理结构并较准确地预测其水动力学特性.结果表明,将浸入边界方法与非平衡滑移速度壁模型相结合的大涡模拟,有望成为数值模拟复杂边界高雷诺数壁湍流的工具.      

一种投影浸入边界法的快速直接求解方法

根据投影浸入边界法分步投影求解的特点,同时针对压力泊松方程离散后的大型稀疏线性方程组是非奇异非对称的特点,结合开源函数库UMFPACK,在传递线性方程组的系数矩阵和右端向量时,采用函数库Eigen将系数矩阵的数据结构改写优化,大大降低了存储空间,实现对高维大型稀疏线性方程组的快速求解,同时求解保持良好的稳定性。本文首先利用一具有解析解的数值算例验证了求解泊松方程数值方法的准确性和网格依赖性,进而利用VC++编写投影浸入边界法的数值计算程序,以单圆柱绕流为基准数值算例,通过与其他文献和实验结果的对比,验证了投影浸入边界法数值计算结果的可靠性,并进一步分析了不同雷诺数下圆柱绕流的流场结构特征和尾涡结构的动态演化过程。     

侧柱与串列双柱绕流之间的干优

本文给出了关于串列双柱与侧柱间流动干扰的实验研究结果。当三个圆柱排成等边三角形关靠得很近时，由于三圆柱间强烈的缝隙流动，大大地改变了绕流其中的串列双圆柱的流态，特别，当三圆柱中心距等于二倍圆柱直径时，在串列双柱的前，后柱之间形成了强烈的偏斜的缝隙流，出现了独特的压力分布以及要比单柱高出三倍以上的旋涡脱落频率。     

基于反馈力浸入边界法模拟复杂动边界流动

浸入边界法是模拟流固耦合的重要数值方法之一。本文采用反馈力浸入边界方法,对旋转圆柱和水轮机活动导叶旋转摆动绕流后的动边界流场进行数值模拟。其中,固体边界采用一系列离散的点近似代替,流体为不可压缩牛顿流体,使用笛卡尔自适应加密网格,利用有限差分法进行求解。固体对流场的作用通过构造适宜的反馈力函数实现。本文首先通过旋转圆柱绕流的计算结果同实验结果进行对比,吻合较好,验证了该计算方法的可靠性。然后针对水电站水力过渡过程中水轮机活动导叶旋转摆动绕流后的动边界流场进行数值模拟,得到导叶动态绕流后的流场分布特性和涡结构的演化特性。     

Asymmetric vortex shedding flow past an inclined flat plate at high incidence

This paper reports an experimental investigation of the vortex shedding wake behind a long flat plate inclined at a small angle of attack to a main flow stream. Detailed velocity fields are obtained with particle-image velocimetry (PIV) at successive phases in a vortex shedding cycle at three angles of attack, α=20°, 25° and 30°, at a Reynolds number Re≈5,300. Coherent patterns and dynamics of the vortices in the wake are revealed by the phase-averaged PIV vectors and derived turbulent properties. A vortex street pattern comprising a train of leading edge vortices alternating with a train of trailing edge vortices is found in the wake. The trailing edge vortex is shed directly from the sharp trailing edge while there are evidences that the formation and shedding of the leading edge vortex involve a more complicated mechanism. The leading edge vortex seems to be shed into the wake from an axial location near the trailing edge. After shedding, the vortices are convected downstream in the wake with a convection speed roughly equal to 0.8 the free-stream velocity. On reaching the same axial location, the trailing edge vortex, as compared to the leading edge vortex, is found to possess a higher peak vorticity level at its centre and induce more intense fluid circulation and Reynolds stresses production around it. It is found that the results at the three angles of attack can be collapsed into similar trends by using the projected plate width as the characteristic length of the flow.     

一种强耦合预估-校正浸入边界法

为克服传统浸入边界法的质量不守恒缺陷,提出了一种用于可压缩流固耦合问题的强耦合预估-校正浸入边界法。通过阐述一般流固耦合系统的矩阵表示,推导了流固耦合系统的强耦合Gauss-Seidel迭代格式,进一步导出预估-校正格式,提出了预估-校正浸入边界法。该方法使用无耦合边界模型对流体进行预估,将流固耦合边界视为自由面,固体原本占据的空间初始化为零质量的单元,允许流体自由穿过耦合边界。对于流体的计算,使用带有minmod限制器的二阶MUSCL有限体积格式和基于Zha-Bilgen分裂的AUSM+-up方法,配合三阶Runge-Kutta格式推进时间步。在校正步骤中,通过一组质量守恒的输运规则来实现输运过程。输运算法可概括为将边界内侧的流体进行标记,根据标记顺序以均匀方式分割和移动流体,产生一个指向边界外侧的流动,最后在边界附近施加速度校正保证无滑移条件。标记和输运算法避免了繁琐的对截断单元的几何处理,确保了算法易于实现。对于固体的计算,分别采用一阶差分格式和隐式动力学有限元格式求解刚体和线弹性体,并利用高斯积分获得固体表面的耦合力。使用预估-校正浸入边界法计算了一维问题和二维问题。在一维活塞问题中,获得了压力分布、相对质量历史和误差曲线,并与其他方法进行了对比。在二维的激波冲击平板问题中,获得了数值模拟纹影和平板结构的挠度历史,并与实验结果进行了对比。研究表明,该方法区别于传统的虚拟网格方法和截断单元方法,能够精确地维持流场的质量守恒并易于实现,且具有一阶收敛精度,能够较准确地预测激波绕射后的流场以及平板在激波作用下的挠度,为开发流固耦合算法提供了一种新的思路。     

Numerical solution of the problem of the mixing of the boundary layers shed from the trailing edge of a wing

During the mixing of viscous incompressible flows with different velocities, in the vicinity of a trailing edge an interaction region with a three-layer structure is formed, similar to that in the case of symmetric shedding with equal velocities. The boundary layers developing on the upper and lower sides of the airfoil form a viscous mixing layer, or vortex sheet, which separates the flows downstream of the trailing edge. The boundary value problem corresponding to the flow in the viscous sublayer in the vicinity of the trailing edge of a flat plate is solved for high Reynolds numbers using an efficient numerical method for solving the equations of asymptotic interaction theory.     

绕振荡水翼流动及其转捩特性的数值计算研究简

通过对比标准k-ω SST 湍流模型和基于标准k-ω SST 湍流模型修正的γ-Re_θ 转捩湍流模型对绕振荡NACA66 水翼流动的数值计算结果与实验结果,对水翼振荡过程的水动力特性和流场结构变化进行了分析研究. 结果表明：与标准k-ω SST 湍流模型的数值计算结果相比,基于标准k-ω SST 湍流模型修正的γ-Re_θ 转捩湍流模型能有效预测绕振荡翼型流场结构和水动力特性,捕捉流场边界层发生的流动分离和转捩现象;绕振荡水翼的流动过程可分为5 个特征阶段,当来流攻角较小时,在水翼前缘发生层流向湍流的转捩现象,水翼动力特征曲线出现变化拐点;随着来流攻角的增大,顺时针尾缘涡逐渐形成并向水翼前缘发展;当攻角较大时,前缘涡分离导致动力失速,水翼的动力特征曲线出现大幅波动;水翼处于顺时针向下旋转阶段,绕水翼的流动状态逐渐由湍流过渡为层流.     

An immersed boundary method for unstructured meshes in depth averaged shallow water models

The representation of geometries as buildings, flood barriers or dikes in free surface flow models implies tedious and time‐consuming operations in order to define accurately the shape of these objects when using a body fitted numerical mesh. The immersed boundary method is an alternative way to define solid bodies inside the computational domain without the need of fitting the mesh boundaries to the shape of the object. In the direct forcing immersed boundary method, a solid body is represented by a grid of Lagrangian markers, which define its shape and which are independent from the fluid Eulerian mesh. This paper presents a new implementation of the immersed boundary method in an unstructured finite volume solver for the 2D shallow water equations. Moving least‐squares is used to transmit information between the grid of Lagrangian markers and the fluid Eulerian mesh. The performance of the proposed implementation is analysed in three test cases involving different flow conditions: the flow around a spur dike, a dam break flow with an isolated obstacle and the flow around an array of obstacles. A very good agreement between the classic body fitted approach and the immersed boundary method was found. The differences between the results obtained with both methods are less relevant than the errors because of the intrinsic shallow water assumptions. Copyright © 2015 John Wiley & Sons, Ltd.     

An Embedding Method for Bluff Body Flows: Interactions of Two Side-by-Side Cylinder Wakes

We introduce a simple method for the numerical simulation of bluff body flows where the solid object is represented by a distributed body force in the Navier–Stokes equations. The body force density is found at every time step to reduce the velocity within the computational cells occupied by the rigid body to a prescribed value. The method combines certain ideas from the immersed boundary method which was developed to treat biofluid mechanical flows and the volume-of-fluid method for simulating flows with fluid–fluid interfaces. The main advantage of this embedding method is that the computations can be effected on a regular Cartesian grid, without the need to fit the grid to the bluff body surfaces. Thus, flow past several complex bodies can be treated as easily as flow past a single body. The method is validated by reproducing well-established results for vortex shedding from a stationary cylinder. The flow past two side-by-side cylinders is then investigated. When the distance between the cylinders is small, they are seen to shed vortices in-phase, whereas for larger distances, the shedding occurs in anti-phase. For intermediate distances, various shedding patterns are observed, including quasi-periodic, asymmetric and chaotic regimes. Mean values and phase portraits associated with the cylinder lift and drag coefficients, as well as spectral analysis of the same data, are used to describe the flow. A transition diagram that can be compared with experiments or models outlines the various dynamical regimes as a function of the distance between the cylinders and the Reynolds number.     

The flexibility effect of a plate according to various angles of attack in a free-stream

This paper presents a computational fluid–structure interaction analysis for a flexible plate in a free-stream to investigate the effects of flexibility and angle of attack on force generation. A Lattice Boltzmann Method with an immersed boundary technique using a direct forcing scheme model of the fluid is coupled to a finite element model with rectangular bending elements. We investigated the effects of various angles of attack of a flexible plate fixed at one of the end edges in a free-stream at a Reynolds number of 5000, which represents the wing flapping condition of insects and small birds in nature. The lift of the flexible plate is maintained at the large angle of attack, whereas the rigid plate shows the largest lift at angles of attack around 30–40° and then drastic reductions in the lift at the large angle of attack. If we consider the efficiency as the lift divided by the drag, the flexible plate shows better efficiency at angles of attack greater than 30° compared to the rigid plate. The better performance of the flexible plate at large angles of attack comes from the deformation of the plate, which produces an interaction between the trailing edge vortex and the short edge vortex. The horseshoe-shaped vortex produced by a large vortex interaction at the trailing edge side has an important role in increasing the lift, and the small projection area due to the deformation reduces the drag. Furthermore, we investigate the role of flexibility on the lift and the drag force of the rectangular plate in a free-stream as the Reynolds number increases. Whenever a large vortex interaction at the trailing edge side is shown, the efficiency of the rectangular plate is improved. Especially, the flexible plate shows better efficiency as the Reynolds number increases regardless of the angle of attack.     

High-order ghost-point immersed boundary method for viscous compressible flows based on summation-by-parts operators

A high-order immersed boundary method is devised for the compressible Navier-Stokes equations by employing high-order summation-by-parts difference operators. The immersed boundaries are treated as sharp interfaces by enforcing the solid wall boundary conditions via flow variables at ghost points. Two different interpolation schemes are tested to compute values at the ghost points and a hybrid treatment is used. The first method provides the bilinearly interpolated flow variables at the image points of the corresponding ghost points and the second method applies the boundary condition at the immersed boundary by using the weighted least squares method with high-order polynomials. The approach is verified and validated for compressible flow past a circular cylinder at moderate Reynolds numbers. The tonal sound generated by vortex shedding from a circular cylinder is also investigated. In order to demonstrate the capability of the solver to handle complex geometries in practical cases, flow in a cross-section of a human upper airway is simulated.