马鞍屋盖表面锥形涡流动模型研究

该文基于大跨平屋盖和马鞍屋盖表面锥形涡流动显示试验,对现有的兰金涡模型进行改进,在其涡核区与势流区之间添加过渡区,从而建立了简化的二维锥形涡流动模型,给出了旋涡上部流速、旋涡内部流线曲率以及屋面涡核吸力之间的定量关系。据此流动模型,分析了旋涡强度的影响因素即为旋涡内部流线曲率和旋涡流速:旋涡内部流线曲率越大,旋涡转速越快,旋涡强度越大。基于流动显示试验和测压试验所得平屋盖以及马鞍屋盖表面各旋涡参数之比,以平屋盖表面锥形涡为基准,量化了各风向下马鞍屋盖表面的锥形涡强度。通过加入考虑旋涡效应对传统准定常理论进行修正,给出了锥形涡涡核吸力(均值、峰值)的计算公式,并将计算值与马鞍屋盖刚性模型风洞测压试验数据进行对比,验证了锥形涡流动模型对于预测马鞍表面锥形涡涡核吸力的有效性。     

基于PIV技术的平屋盖表面分离泡流动结构研究

该文利用粒子图像测速技术,通过风洞流场显示试验,观察了大跨平屋盖表面的分离泡现象,给出了不同来流工况下多个可视化平面的旋涡流线和涡量场分布。试验结果表明,当风向垂直于平屋盖迎风前缘时,屋盖顶面将出现典型的分离泡现象,涡量场的负向峰值出现在迎风前缘处,屋面风压力随离迎风前缘距离的增加而减小。均匀流场下流动将在迎风边缘产生分离而后再附,再附长度近乎横跨整个屋盖;而湍流场中的小尺度湍流促使分离剪切层较早地再附形成分离泡,且湍流度越大,旋涡再附长度越短。运用FLUENT模拟了平屋盖表面的分离泡,与流动显示试验结果吻合较好。通过多个可视化平面的综合分析得到分离泡的三维形态特征,建立了旋涡的演化、涡核位置与建筑物表面压力分布的内在联系,获得了若干有价值的结论。     

马鞍屋盖表面分离泡流动形态研究

利用粒子图像测速技术,通过风洞流场显示试验,给出了均匀流和格栅紊流两种工况下马鞍屋盖迎风高点和迎风低点处旋涡的截面形态特征以及时均、瞬时流线和涡量场分布,比较了两种工况下旋涡形态特征的异同,总结了来流在屋面上的速度分布规律。试验结果表明:当风向垂直于迎风前缘时,屋面迎风高点至迎风中点区域将出现典型的分离泡现象;而在迎风中点至低点区域,由于马鞍屋面为正向坡度,无法形成分离泡。均匀流场下分离泡再附长度近乎横跨整个屋盖;而湍流场中分离泡涡核离迎风前缘较近,且旋涡横截面积显著小于均匀流工况。结合风洞测压试验,分析了不同迎风高度处马鞍屋盖表面的脉动风压谱特性,明确了谱能量与旋涡运动或湍流尺度之间的演变关系。利用多个可视化平面的综合布置最终揭示出分离泡的三维形态特征,据此将马鞍屋盖划分为4个区域并提出合理的分区体型系数,为大跨马鞍屋盖抗风设计提供参考。     

锥形涡及其诱导下的马鞍屋盖表面风荷载

通过刚性模型风洞测压试验,对锥形涡及其诱导下的菱形马鞍屋盖表面风荷载进行了系统研究:⒈分析了锥形涡诱导下的屋盖表面平均、脉动风压分布特点;通过脉动风压谱,着重分析了马鞍迎风前缘(易损区域)的平均、脉动风压分布机理。⒉通过风压信息确定了锥形涡的特征参数(涡核位置、作用范围),给出了定量结果;分析了锥形涡涡轴的左右摇摆运动,分别从定性和定量方面建立了锥形涡涡轴的运动与屋面吸力脉动之间的关系。⒊用测点风压时程的三、四阶矩对风压的非高斯特性进行描述,给出划分高斯、非高斯区的标准,并在此基础上对马鞍屋盖表面进行了分区;分析了锥形涡作用下屋盖表面风压非高斯特性的产生机理。     

马鞍屋盖表面面积平均风压特性研究

为完善我国建筑荷载规范中有关围护结构和覆面材料的设计,本文提出了面积平均风压的概念。通过刚性模型风洞测压试验,研究了马鞍屋盖表面的面积平均风压特性。⒈ 以锥形涡为引,确定各矢跨比马鞍屋盖表面出现锥形涡的来流工况,以涡核吸力和涡核互相关性为依据,确定了马鞍屋盖的最不利风向为沿高点连线方向。⒉ 在一定附属区域内对单点风压进行集合,得到面积平均风压（时程）;在马鞍屋盖表面实行各种分区,探讨了最不利风向下各面积平均风压时程的峰值吸力随附属面积的变化趋势,得到了相应的拟合公式;通过各区域内单点风压时程的累积概率分布,验证了该拟合公式的准确性与有效性;根据偏度和峰度值大小,分析了面积平均风压时程的非高斯特性,并与单点时程的风压特性进行了比较,结果表明面积平均风压时程曲线中风压脉动分布均匀,高频脉动减少,因而其非高斯特性并不如单点突出,且普遍偏向高斯分布     

典型形式大跨度屋盖风雪漂移的数值模拟

近年来,因积雪的不均匀分布造成大跨度屋盖坍塌的事故已屡见不鲜,这主要是由于屋盖上的积雪在风力作用下发生复杂的风雪运动,从而造成大跨度屋盖表面积雪的不均匀分布,正确预测大跨度屋盖上的积雪分布有着非常重要的意义。基于Mixture多相流模型和k-kl-ω模型对几种典型形式大跨度屋盖的风雪漂移进行了数值模拟,探讨了不同形式屋盖的积雪分布规律,研究了风向角、矢跨比对屋面积雪分布的影响,并将平屋盖和拱形屋盖数值模拟结果与规范进行了对比分析。     

扰流板对分离泡诱导下平屋盖表面风荷载的影响研究

通过风洞测压试验,研究对比了设置扰流板前后,分离泡诱导下平屋盖表面风压特性的变化;基于此,分析了扰流板作用机理,并给出了扰流板的推荐倾角.研究了平屋盖表面风压分布及统计值的变化,结果表明:设置扰流板后,除迎风前缘附近,屋面平均和脉动风压均减小;分离区内风压偏度和峰度值显著降低,非高斯脉动特性减弱.探讨了扰流板对于顺风向...     

下击暴流作用下菱形马鞍面屋盖风压特性

设计制作了菱形马鞍面大跨屋盖结构缩尺模型,通过风洞试验探究了下击暴流作用下菱形马鞍面屋盖表面风压特性.探究了冲击射流装置到模型的径向距离和风向角对屋盖上不同区域平均风压的影响,试验结果表明:最大分区平均风压系数出现在径向距离为1.25Djet(Djet为喷口直径)、风向角为0°的工况;来流方向上屋盖檐口的3个区域平均风...     

大跨平屋盖风荷载特性及风压预测研究

在大气边界层风洞中开展了大跨平屋盖结构刚性模型试验,获得了屋盖表面测点的风压时程,分析了典型风向下屋盖表面平均风压与脉动风压特性。结合本征正交分解技术(POD)与BP神经网络法,提出了一种可用于大跨结构进行空间插值的机器学习法—POD-BPNN法,实现了对风压的高效预测。预测的平均风压系数、脉动风压系数、脉动风压的时域与频域特性均与风洞试验值相吻合。表明运用POD-BPNN方法预测大跨结构表面风压是可行的。     

基于大涡模拟的大气边界层湍流强度对低矮房屋风荷载特性影响研究

对德州理工大学(Texas tech university,TTU)低矮房屋标准模型,以已有现场实测以及缩尺模型风洞实验数据为验证对比,基于大涡模拟(Large-eddy simulation,LES)方法研究了大气边界层湍流强度对低矮房屋风荷载特征的影响机理。采用CDRFG (Consistent discretizing random flow generation) 人工合成湍流方法生成大气边界层湍流,研究了来流湍流度对低矮建筑表面的平均、脉动以及极小值风压分布以及风压非高斯特性的影响,并利用LES能提供非常场流动全流域信息的优势,结合瞬态湍流场结构对大气边界层湍流对低矮房屋风荷载特征的影响机理进行了阐释。结果表明:LES数值模拟得到的平均、脉动及极小值风压系数与实验以及实测结果一致,平均风压结果包络在实测误差范围以内,极小值风压系数最大误差小于10%,脉动风压系数最大误差小于20%且误差区域较小。在来流湍流度增大的过程中,低矮房屋屋面平均风压系数变化较小,脉动风压系数呈显著的线性增加;极小值风压系数变化规律相对复杂,呈现出非线性减小的趋势,风压系数极小值可达?5.0;屋面涡脱强度逐渐被抑制,锥形涡迹线与屋面迎风前缘的夹角由14.4°下降至8.7°。屋面风压非高斯特性主要与屋面形成的涡旋结构相关,表现出典型的右偏软化非高斯过程,且随着来流湍流度的增加风压非高斯特性逐渐减弱。从流场的角度来看,湍流度的增加抑制屋面迎风前缘柱状涡以及锥形涡的形成,加快流动分离的再附,减少分离泡尺度,同时提高了屋盖周围的湍流高频能量成分,从而使脉动风压增加,极小值风压减小以及风压非高斯特性减弱。该研究阐明了大气边界层湍流对低矮房屋风荷载特性的影响机理,有助于进一步理解低矮房屋风致破坏机理,并且为低矮房屋的抗风设计及抗风性能优化提供重要参考。     

Wind pressure features of large-span flat roof in different wind fields induced by conical vortex

The wind pressure features on a large-span flat roof in uniform flow field and turbulent field induced by conical vortex were studied, through wind tunnel tests. From the comparison of the mean and fluctuating wind pressure distributions on a flat roof in different wind fields induced by conical vortex, results indicate that the mean suction dominates in the smooth flow, whereas the fluctuating suction is more obvious in the turbulent flow. The probability density function for the pressure fluctuations under different approaching flows is analyzed. The two-peaked distribution, peculiar to turbulent flow field, is observed on the curve of probability density. The fluctuating pressures at reattachment points are larger under the turbulent flow. This indicates a more intense reattachment, which may cause overturning moment for roof-mounted items. Point vortex, RanKine vortex, and simplified Cook expression are applied to fit the pressure profiles beneath conical vortices, respectively. The results have shown that the RanKine vortex model and simplified Cook expression were applicable to forecast the wind pressure profiles beneath conical vortices, while point vortex underestimated the real wind suction. The wind pressure distributions in turbulent fields induced by different wind angles were contrasted, when the approaching flow is along the diagonal of the roof, the intensity of the vortex pairs is almost equal, with obvious reattachment. When the approaching flow deviate from the diagonal of the roof, the lateral turbulent component spins the vortex more quickly; this induces larger mean suctions beneath windward vortices. Smaller suctions are observed beneath the leeward vortex, due to less vorticity being converted to vortex motion from the freestream.     

Wind loads on flat roofs of low-rise buildings with rounded leading edge

Using wind tunnel experiments, wind loads on the flat roofs of low-rise buildings with rounded leading edges were investigated under the effects of separation bubbles and conical vortices. First, pressure distributions on the flat roofs were presented. Second, total uplift forces and overturning moments were calculated by the introduction of area-averaged pressures. Third, eigenvalues and eigenvectors of fluctuating pressure were obtained by proper orthogonal decomposition. Finally, wind pressure characteristics at the rounded edges were examined. The results indicate that after the use of a rounded leading edge, suctions induced by separation bubble and conical vortices increase near the chamfer, decrease beneath the vortices, and stay invariant far from the leading edge. The area involved and the energy of pressure fluctuation decrease, with the largest decrease occurring on a flat roof with a larger fillet radius under conical vortices. In addition, the maximum decreases in peak total uplift forces and overturning moments reach 11% and 62%, respectively, and appear at skewed flows. The maximum suction and low-frequency spectral peak of pressure fluctuations at the rounded leading edge may exceed those at the neighboring region. As the fillet radius increases, the maximum suction and low-frequency pressure fluctuation energy at the chamfer decrease.     

基于非概率集合理论的屋面板风致疲劳寿命估计

现有疲劳分析中,通常将结构材料参数、几何尺寸等定义为确定性参数;实际结构中,相关参数均为有界但不确定变量,如按确定性参数估计结构的疲劳寿命是偏于不安全的。该文将结构体系中不确定参数定义为区间变量,在线性疲劳损伤累积理论基础上,提出一种仅需一次动力响应分析即可计算不确定结构在动力荷载作用下疲劳损伤的新方法。该方法将金属屋面板弹性模量和屋面板板厚等由于施工误差等因素引起的不确定参数定义为区间变量,通过摄动法和区间动力响应分析,计算屋面板在脉动风荷载作用下的应力响应区间;结合屋面板材料的S-N曲线,采用修正Miner疲劳线性累积准则对屋面板的疲劳损伤和寿命区间进行估计。结果表明:该文方法可有效计算考虑结构参数不确定条件下金属屋面板的疲劳损伤和寿命区间;与顶点法比较,该文方法仅需一次动力响应分析就可计算金属屋面板风致疲劳损伤和寿命区间。     

Simulating three-dimensional vortex motion by a vortex blob method

A three-dimensional vortex blob method was applied to calculate several vortex motions: the deformation of pseudo-elliptic vortex rings, the jet issuing from the pseudo-elliptic nozzle into flow of uniform velocity, the unsteady separated flow around a circular disk with an angle of attack, and the interaction of several vortex rings which approximately reproduced the Kolmogorov spectrum. In the first three cases, the viscous diffusion of vorticity was included. The pseudo-elliptic vortex rings experienced axis switching and split into a few deformed vortex rings. Rolling-up vortices in the pseudo-elliptic jet had a symmetric arrangement in the minor-axis plane and an antisymmetric arrangement in the major-axis plane in the developing region; further downstream, the vortices were arranged antisymmetrically in both planes. The wake behind the disk normal to the main flow reproduced the spiral and columnar modes of instability. A problem in the three-dimensional vortex method is that vorticity tends to diverge at a stage of evolution of the vortex motions. An approximate method of avoiding the divergence of vorticity is proposed.     

大跨度平屋面的风振响应及风振系数

本文在有限元分析的基础上建立了大跨度平屋面结构在风荷载作用下的风振响应谱分析方法,并采用Davenport谱和由风洞试验得到的屋盖表面的平均风压分布系数计算了屋面的风振响应及风振系数。文中还深入探讨了屋面刚度、来流风速及风向等参数对大跨度平屋面竖向风振响应及风振系数的影响。计算表明:① 大跨度平屋面的竖向风振响应主要是由第一振型所支配,高阶振型对屋面板竖向风振响应的影响很小;② 屋面刚度及来流风速对大跨度平屋面的竖向风振响应影响比较大,但对位移风振系数的影响不太明显;③ 在工程设计中,建议采用位移风振系数来计算大跨度平屋面的等效静力风荷载。     

A flat plate in a rotating,stratified flow

Summary The boundary layer over a flat plate of semi-infinite extent in a stratified and rotating flow grows forward from the trailing edge, and is characterized by an intrinsic length scaleL, which represents the distance from the trailing edge at which vortex stretching becomes just as important in the boundary layer as baroclinic vorticity production. Near the trailing edge, the layer is essentially the layer in a purely stratified flow; far upstream (manyL), it is an Ekman layer. The boundary layer entrains no fluid, but induces at its edge a transverse velocity component which drives an higher-order streamwise outer flow. If the flow is bounded above and below by horizontal planes, the Wiener-Hopf solution for this outer flow indicates that the disturbance decays rapidly downstream, but the transverse velocity component is non-zero far upstream.