质量比对圆柱涡激特性的影响研究

采用有限体积法对不同质量比圆柱在限制流向及不限制流向下的涡激振动进行了研究。圆柱涡激振动系统简化为质量-弹簧-阻尼模型,引入雷诺平均应力模型求解不可压缩粘性Navier-Stokes方程,结合SST  湍流模型对限制流向和不限制流向下圆柱涡激振动进行了数值模拟。研究发现：限制流向和不限制流向时圆柱涡激振动横向振幅均出现了初始激励分支和下端分支, 不限制流向质量比2.0时还出现了超上端分支,其横向振幅最大值为1.05D,是限制流向工况的1.81倍,质量比越大两者相差越小;限制流向和不限制流向两种工况下圆柱涡激振动均发现频率锁定现象,但锁定区间不同;质量比大小对圆柱涡激振动锁定区间也有影响;最后对不同质量比下圆柱涡激振动轨迹进行了讨论分析。     

CSM在基于Lamb波的结构损伤检测中的应用

在基于Lamb波的结构损伤检测中往往不可避免地遇到结构中多个损伤引起的反射波信号相干问题而对检测精度造成影响。考虑到阵列信号处理中的空间谱估计可以对信号源进行辨别和定位,尤其是在信号的波达方向(Direction of Arrival, DOA)估计上的优越性,将其应用于基于Lamb波的结构损伤检测中以获得准确的结构损伤信息。于是,该文采用空间谱估计中相干信号子空间方法(Coherent Signal Subspace Method, CSM)解决了相干信号问题。以一个损伤结构的仿真分析为例,验证了CSM方法在基于Lamb波的结构损伤检测中处理相干信号的可行性。结果表明,无论是结构损伤和边界反射引起的信号相干或两个结构损伤引起的信号相干,依然可以通过该方法获得准确的结构损伤信息。     

基于离散点涡的双自由度涡激振动尾流振子模型研究

建立了一种新的预报双自由度圆柱涡激振动响应的尾流振子模型,提出了以近壁点涡强度为变量描述尾流振子方程.通过势流理论推导出了结构所受流体流向和横向水动力与点涡强度的量化关系,从而得到了结构振子和流体振子的耦合方程组.使用该模型对圆柱的双自由度涡激振动问题进行了数值计算,得到了结构流向和横向的响应振幅及频率的变化规律.通过与实验结果进行对比,表明模型预测的涡激共振区、锁定频率和“8”字形运动轨迹等结果与实验观测结果基本一致.     

行波壁抑制圆柱涡激振动的数值模拟研究

采用CFD数值模拟方法完成了圆柱绕流-涡激振动-行波壁流动控制全过程的数值模拟,重点研究行波壁流动控制方法对低雷诺数下两自由度弹性支撑单圆柱涡激振动的抑制作用。详细分析各阶段圆柱横向和流向位移、质心运动轨迹、升力和阻力系数等随频率比的变化。结果表明:行波壁圆柱的波谷处可以产生一系列稳定的随行波壁运动的小尺度旋涡,有效抑制圆柱表面分离涡的产生,达到消除圆柱绕流尾迹和抑制涡激振动的目的;在计算初始和中途启动的行波壁流动控制方法显著抑制了圆柱横向和流向振动、降低了圆柱升力系数脉动值和阻力系数均值,但阻力系数脉动值则明显增大。     

微型三角楔超声速绕流特性的研究

基于大涡模拟(LES)方法,结合WENO格式与自适应网格加密(AMR)技术及沉浸边界法(IBM),对来流马赫数为Ma =2.5条件下的平板上微型三角楔绕流流场进行了数值模拟。数值模拟表明微型三角楔涡流发生器可以显著改变超声速流体边界层结构。计算结果清晰地显示了三角楔上游分离区的流场结构和下游各涡的流态,同时计算表明,微楔对边界层控制过程中,其下游的流向涡对与涡环结构都起了重要作用,并对其作用过程进行了讨论。数值计算与相关实验结果相符,且提供了流场的重要细节,揭示了微楔的控制机理,可为超声速边界层控制研究提供重要支持。     

基于小波变换的顶张力立管涡激振动规律实验研究

为研究顶张力立管在海流作用下的涡激振动规律,在大型波浪流水槽中进行涡激振动模型实验。实验中立管竖直固定于实验支架上,外部承受不同速度的流体作用,上端施加变化的顶张力。立管模型上均匀布置六个测点,根据每个测点布置的两个应变计,分别测得来流向和横向两个方向振动响应。通过小波变换方法对实验数据进行时间频域分析,得到立管涡激振动频率和振动幅值以及来流向与横向的耦合振动规律,考察顶张力变化对立管自振频率以及涡激振动的影响。     

平台运动与管内流动联合作用下悬垂立管动力响应特性研究

悬垂立管因其经济性被广泛应用在深海油气开发中。悬垂立管在顶端平台运动和内流共同作用下,在顺流向(IL向)会产生动力响应,并与周围流体间形成相对振荡来流。这种相对振荡来流极有可能诱发悬垂立管产生横流向(CF向)的涡激振动(Vortex-induced Vibration,VIV)。为了研究悬垂立管在平台运动和内流联合作用下悬垂立管涡激振动响应特性,开展了悬垂立管模型试验。通过试验和数值模拟的方法,分析了IL向应变、流速、KC数、泻涡频率等分布规律。结果表明:大管径悬垂立管在振荡流场和内流共同作用下,可以产生涡激振动。该试验内流速度区间下,内流作用对于悬垂立管固有频率,KC数和泻涡频率幅值分布,以及涡激振动影响不大。顶部平台运动诱发的涡激振动具有不稳定性。     

微涡流发生器对激波边界层作用诱导的流体分离控制

基于大涡模拟(Large Eddy Simulation)方法,结合高阶TCD/WENO混合格式,对2.5Ma超声速激波/边界层干扰诱导的微楔和微叶片两种微涡流发生器控制进行了数值模拟。数值结果表明:从边界层厚度与分离区大小及结构变化均说明两种涡流发生器对激波边界层分离起到明显抑制作用。对于微楔式涡流发生器,入射激波对微楔尾涡压缩作用明显,使流向涡对的卷吸加强,从而增加边界层内外流体能量交换。微叶片式涡流发生器的控制机理与微楔相似,但其涡对在尾部较远处易破裂,会影响内外流体之间的能量交换。     

高雷诺数下圆柱顺流向和横向涡激振动分析

本文利用CFD方法,研究了较高雷诺数下圆柱流向与横向耦合涡激振动特性。利用FLUENT软件求解粘性Navier-Stokes方程、圆柱涡激振动的结构动力响应方程,运用动网格技术,实现流固耦合,对圆柱进行了单自由度和两自由度涡激振动的数值模拟,得到了雷诺数为 范围内的圆柱涡激振动的升力系数、阻力系数、振幅比及频率比随约化速度变化的规律,捕捉到涡激流固耦合振动的“锁定”“相位开关”等现象,结果表明在此雷诺数范围内锁定区域对应的折减速度范围为Ur=3~7.5。对比单自由度及两自由度的模拟结果,表明在低质量比情况下,流向的振动会对横向振动产生影响。     

两自由度椭圆柱体涡激振动的数值模拟

摘要：利用作者编制的基于松耦合方法求解气动弹性问题的程序,对高雷诺数条件下椭圆柱体的涡激振动进行了数值模拟。椭圆柱体的长短轴比接近2,来流和长轴垂直。数值结果表明：尽管椭圆柱体在横向和流向的动力参数不同,结构振动仍以横向振动为主;和圆柱以及方柱一样,两自由度椭圆柱体的涡激振动同样存在流向锁定和横向锁定,并且随着折减风速的增大,旋涡脱落模式从异相2S变化为同相2S。     

逆压梯度下近壁湍流的喷射和扫掠

采用Fourier谱展开和紧致有限差分格式，选用两组共振三波为相干结构的初值，计算了其在零压和逆压梯度作用下的演化。对演化后期流场的2，4象限的运动进行了详细的分析。结果发现，在逆压梯度下，扫掠对雷诺应力的贡献要强于喷射。无论是在零压梯度还是逆压梯度下，uv和u2在法向的输运主要是靠Q2和Q4这两种运动来完成的。零压梯度下喷射部分对输运的贡献大于扫掠的部分。而在逆压梯度下喷射部分对输运的贡献明显减少，扫掠的作用要强于喷射。     

Gradient design of metal hollow sphere (MHS) foams with density gradients

Metal hollow sphere (MHS) structures with a density gradient have attracted increasing attention in the effort to pursue improved energy absorption properties. In this paper, dynamic crushing of MHS structures of different gradients are discussed, with the gradients being received by stacks of hollow spheres of the same external diameter but different wall thicknesses in the crushing direction. Based on the dynamic performance of MHS structures with uniform density, a crude semi-empirical model is developed for the design of MHS structures in terms of gradient selections for energy absorption and protection against impact. Following this, dynamic responses of density graded MHS foams are comparatively analyzed using explicit finite element simulation and the proposed formula. Results show that the simple semi-empirical model can predict the response of density gradient MHS foams and is ready-to-use in the gradient design of MHS structures.     

Effect of wall-temperature distribution on heat transfer in centrifugal flow in the gap between parallel rotating disks

The effect of a radial change in the wall temperature on heat transfer in centrifugal turbulent flow in a gap between parallel rotating disks is investigated. Cases of positive, approximately constant, and negative radial gradients of the disk temperature are considered. The results of calculating by the integral method are in good agreement with the known experiments. It is shown that the change in the wall temperature has an insignificant effect in the region of the source and a determining effect in the Ekman layers, finally giving rise to a zone with return heat flux at the periphery of the cavity with negative and approximately constant gradients of the disk temperature.     

Continuous precipitation in Cu-rich Cu-Ti binary and Cu-Ti-Al ternary alloys

An investigation has been carried out to study precipitation in a Cu-4% Ti binary and a Cu-2.1% Ti-2.4% Al ternary alloys which develop modulated microstructures upon ageing. Precipitate-free zones near the grain boundaries were observed on ageing immediately below a critical temperature representing the transition from the coherent to the incoherent region of the coherent phase diagram. In addition, a pronounced tendency for elastic interaction between the precipitate particles was detected, giving rise to aligned groups of precipitates containing a maximum of eight or nine particles in each group. At least near this critical temperature, decomposition occurs in the metastable region of the coherent phase diagram rather than in the spinodal region. The formation of the modulated structures in these alloys is, therefore, considered to arise not necessarily through spinodal decomposition. Continuous precipitation in a Cu-2.1% Ti-5% Al alloy is shown to result in spherical particles with no alignment. The effect of Al addition on the coherent solvus temperature is discussed in terms of a vacancy solute interaction mechanism.     

Micropolar fluid jet impingement on a curved surface

This paper deals with an analysis of a normally impinging micropolar fluid jet on a curved surface. The flow near the stagnation point in the impingement region is divided into inviscid and viscous flow regions. The inviscid flow solution is governed by Euler's equations of motion expressed in curvilinear coordinate system. The viscous flow solution is governed by the zeroth and the first order boundary layer equations. These boundary layer equations are solved by assuming power series expansions for both velocity and microrotation fields which give rise to two systems of ordinary coupled differential equations. The effects of surface curvature and material parameters on boundary layer characteristics have been studied and presented graphically. The gradients of zeroth order velocity and microrotation at the wall decrease and the zeroth order displacement and momentum thicknesses increase with decrease in the value of surface curvature. The reduction in curvature results in the reduction in the gradients of first order velocity and microrotation at the wall as well as first order displacement and momentum thicknesses.     

棱纹面与洛仑兹力对湍流边界层特性影响的比较

用氢气泡技术对棱纹面平板以及施加行渡式洛仑兹力的平板水边界层开展流动显示实验研究，探讨棱纹面与行渡式洛仑兹力影响湍流边界层相干结构的作用机理．实验结果表明，棱纹面与洛仑兹力促进了边界层由层流向湍流的转捩；而对于湍流边界层，在一定的无量纲参数范围内棱纹面与行渡式洛仑兹力都能实现边界层的局部减阻，相应地湍流边界层粘性底层低速条带平均间距有所增加．     

Multiscale simulations of carbon nanotube nucleation and growth: mesoscopic continuum calculations

As part of a focused computational effort on the multiscale simulations of carbon nanotube nucleation and growth, we have developed computer programs for coupled heat and mass flow in one and two dimensions. In the tip-growth mode, the sample is divided into three main regions, each of which can be further subdivided as required. In region 1, carbon is supplied to the catalytic particle from an ambient gas of carbon-containing compounds. The chemistry and thermodynamics of the decomposition of these compounds can be included in region 1, but the capability has not yet been implemented. The carbon diffuses through the catalytic particle in region 2 under concentration and temperature gradients and with a diffusion coefficient that can depend on both concentration and temperature. Region 3 consists of the interfacial region between the catalytic particle and the growing nanotube. Results to date demonstrate the key roles played by the size and shape of the catalytic particle in conjunction with the concentration and temperature gradients at the gas/solid interface and in region 2. Results also suggest how the growth of a single wall may interfere with, but not necessarily prevent, the growth of additional walls in a multi-walled nanotube. Again, the carbon concentration profile in the catalytic particle at the different growth sites is a key factor.     

Behaviour analysis of numerical solutions and simulation of coherent structures in compressible mixing layers

For understanding the correctness of simulations the behaviour of numerical solutions is analysed. In order to improve the accuracy of solutions three methods are presented. The method with GVC (group velocity control) is used to simulate coherent structures in compressible mixing layers. The effect of initial conditions for the mixing layer with convective Mach number 0.8 on coherent structures is discussed. For the given initial conditions two types of coherent structures in the mixing layer are obtained.     

Nanopore gradients on porous aluminum oxide generated by nonuniform anodization of aluminum

A method for surface engineering of structural gradients with nanopore topography using the self-ordering process based on electrochemical anodization of aluminum is described. A distinct anodization condition with an asymmetrically distributed electric field at the electrolyte/aluminum interface is created by nonparallel arrangement between electrodes (tilted by 45°) in an electrochemical cell. The anodic aluminum oxide (AAO) porous surfaces with ordered nanopore structures with gradual and continuous change of pore diameters from 80 to 300 nm across an area of 0.5-1 cm were fabricated by this anodization using two common electrolytes, oxalic acid (0.3 M) and phosphoric acid (0.3 M). The formation of pore gradients of AAO is explained by asymmetric and gradual distribution of the current density and temperature variation generated on the surface of Al during the anodization process. Optical and wetting gradients of prepared pore structures were confirmed by reflective interferometric spectroscopy and contact angle measurements showing the ability of this method to generate porous surfaces with multifunctional gradients (structural, optical, wetting). The study of influence of pore structures on cell growth using the culture of neuroblastoma cells reveals biological relevance of nanopore gradients and the potential to be applied as the platform for spatially controllable cell growth and cell differentiation.     

Accurate determination of magnetic field gradients from four pointvector measurements. I. Use of natural constraints on vector dataobtained from a single spinning spacecraft

Cluster introduces a new generation of spacecraft that will measure the spatial gradients of the magnetic field in the Earth's magnetosphere. As gradients require knowledge of differences, small errors resulting from an inadequate knowledge of the orientations, zero levels and the scale factors of the magnetometer sensors affect the calculation of field gradients disproportionately and must be removed with high accuracy. We show that twelve calibration parameters are required for each of the spacecraft (for a total of 48 for the four spacecraft) to correctly infer the measured magnetic fields at each of the spacecraft. By application of a Fourier transform technique, some of the parameters can be recovered. We show that errors in eight of the twelve calibration parameters generate coherent monochromatic signals at the first and the second harmonics of the spin frequency in the despun data. We relate the real and the imaginary parts of the monochromatic signals to the eight calibration parameters. We then present a least squares scheme that improves the eight calibration parameters by iteration until the power of the coherent signal superimposed above the broad-band background is minimized