DISCUSSION ON THE WAKE VORTEX STRUCTURE OF A HIGH SPEED TRAIN BY VORTEX IDENTIFICATION METHODS
PanYongchen1, YaoJianwei2, LiuTao3, LiChangfeng3,*, 1 Postgraduate Department, China Academy of Railway Sciences, Beijing 100081, China2 Railway Science & Technology Research & Development Center, China Academy of Railway Sciences, Beijing 100081, China;3 School of Energy and Power Engineering, Jiangsu University, Zhenjiang 212013, Jiangsu, China 中图分类号:O357.52,U266 文献标识码:A
关键词:高速列车;湍流尾流;涡旋结构;涡旋识别方法 Abstract A flow field around a 1/30th-scale and simplified model of high speed train (HST) has been numerically calculated by the improved delayed detached eddy simulation, and the vortex structure in the near wake detailedly discussed as a focus. According to different vortex identification methods, it can be observed for the wake vortex structure that powerful vortices with high vorticity magnitude mostly appear in the vicinity of the tail; however, there are stable vortices with lower vorticity widespread in the near wake region. Based mainly on the findings and the newly-proposed definition of a vortex and physical meaning, there are conclusions given as follows. Shear deformation and high vorticity diffused play significant roles in forming those energetic vortices, due to boundary layers separated from the streamlined tail. And turbulent eddies have to be rotated and strained by the strong shears, thus resulting in prominent turbulent characteristics of the local complex flow. On the other hand, though strength of the vortices evidently drops, vortical vorticity is dominant inside the streamwise vortex cores when the strong shear strains rapidly decay. Under the circumstances, fluid particles rotate round the cores that get closer to the ground, and thus the interaction between the vortices and the ground becomes a dominant mechanism. The vortices have to be diminished at relatively low rate, but considering turbulence production, the flow mechanism can play an important role in self sustainment of turbulent eddies. As a result, the vortex structure is able to stably be in the wake flow.
Keywords:high speed train;turbulent wake;vortex structure;vortex identification method -->0 PDF (15291KB)元数据多维度评价相关文章收藏文章 本文引用格式导出EndNoteRisBibtex收藏本文--> 潘永琛, 姚建伟, 刘涛, 李昌烽. 基于涡旋识别方法的高速列车尾涡结构的讨论[J]. 力学学报, 2018, 50(3): 667-676 https://doi.org/10.6052/0459-1879-17-383 PanYongchen, YaoJianwei, LiuTao, LiChangfeng. DISCUSSION ON THE WAKE VORTEX STRUCTURE OF A HIGH SPEED TRAIN BY VORTEX IDENTIFICATION METHODS[J]. Chinese Journal of Theoretical and Applied Mechanics, 2018, 50(3): 667-676 https://doi.org/10.6052/0459-1879-17-383
用于模拟的是缩尺比例为1:30的CRH380A高速动车组模型,由头车和尾车组成,长度约为 ( 为模型高度). 按照空气动力学模型的一般简化处理方法,不考虑受电弓部件以及车窗和车门等细节. 除此之外,还进一步对模型底部转向架位置作了理想化处理. 并不像Muld等[16] (用形状简单的外罩包裹转向架的复杂结构),这里将转向架删除,并用与边缘重合的盖板密封原来转向架所在位置的空腔,如图1所示. 通过对模型底部结构的简化,一定程度上降低了局部流动可能导致的计算上的不稳定性. 列车模型在整体上保留了对列车车体表面边界层以及近尾流区流动拟序结构的生成和发展起重要作用的基本几何特征,即较大的长细比、小高宽比以及头车和尾车的流线型头部. 显示原图|下载原图ZIP|生成PPT 图1基于CRH380A高速列车的简化模型 -->Fig. 1Simplified model based on the CRH380A HST -->
计算网格主要由近壁面区域的棱柱层(prism layer)以及远离壁面区域的六面体结构网格(hexahedral mesh)构成,两者通过切割体网格(trimmed mesh)过渡连接. 车体表面设置12层边界层网格(即棱柱层),其中第一层边界层网格单元的无量纲厚度 的量级 为1;壁面边界层网格中,其他两个方向的网格间距约为0.017 ,该值约是 的50倍. 采用与Muld等[17]相似的方法,对车身周围以及尾流区进行网格加密;共细化了5个区域的网格,为了能够获取影响 尾车气动特性的大尺度涡旋结构,其中有3个网格细化区设置在近尾流区域,如图3所示. 在此基础上进行网格无关性验证. 验证所用模型的长度约为22 ,其他几何尺寸不变,流动条件如前所述. 3种网格加密方案(即细密网格FM、中等网格MM以及粗糙网格CM)所对应的网格单元总数分别约为17.9×106, 10.5×106和6.3×106. 显示原图|下载原图ZIP|生成PPT 图3网格加密区示意图及局部放大 -->Fig. 3Schematic view and close-up of the zones with grids refined -->
图4给出了利用3种网格得到的近尾流区中流 向速度 和湍动能(turbulent kinetic energy)时均分布( ;式中, 为张量下标,上标撇号表示相应变量的脉动). 如图4所示,MM曲线与FM曲线在速度和湍动能分布上总是能够保持一致的趋势,并彼此吻合得较好. 而图4(b)中的CM速度曲线有些偏 离其他两条曲线,并且CM湍动能曲线沿车长方向的变化趋势表现出明显的不同. 根据验证结果划分计算网格,用于研究讨论尾流区涡旋结构. 其中,加密区网格单元尺寸控制在0.02H~0.08 ,网格总数约为9.3×106. 显示原图|下载原图ZIP|生成PPT 图4利用3种网格计算得到的近尾流区流向速度和湍动能沿车长方向的时均分布 -->Fig. 4Time-averaged profiles of streamwise velocity and as a function of the streamwise distance away from the tail in the near wakes of the trains with the three different grids -->
1.5 数值结果的验证
列车风是高速列车尾流区域中典型的流动现象,其发展与涡旋结构有紧密的联系. 这里,鉴于模拟中高速列车模型是静止的,将列车风 速度 定义为水平面( - 平面)上速度分量的合成,即 $u_{\rm S} = \sqrt {(U_\infty - u_x )^2 + u_y ^2} (2)$ 式中, 和 分别为列车周围流场在流向( )和展向( )方向上的瞬时流动速度. 图5给出了数值模拟与试验得到的列车风速度曲线图. 图中 为列车模型的宽度,宽高比 ; 对应尾车鼻端, 为列车模型和尾流的展向对称中心, 为模型底面. 在Bell等[12]的风洞试验中,封闭测试段安装有分流平台,以削弱地面边界层效应;相似地,目前模拟中计算域地面采用无滑移运动边界条件,同样可以显著减小该效应. 显示原图|下载原图ZIP|生成PPT 图5尾流区中列车风速度沿流向的变化曲线 -->Fig. 5Time-averaged slipstream as a function of the streamwise distance from the tail -->
涡量和变形对定义流动中的涡旋都是很重要的. 流体微团受到强剪切的作用时,变形是显著的,但因此产生的涡量可能并不与涡旋运动直接相关,涡旋涡量会下降. 正如Liu等[25]所指出的,当旋转很强时,涡量可能会很小;而旋转很弱甚至没有时,涡量却可能很大. 基于此,可以预料,头部表面的高涡量层与局部流体微团的剪切变形有密切关系,某种程度上暗示出流线型几何外形延迟了由车体表面发展出的边界层的流动分离,而这也将影响到流向尾涡的形成和脱落.随着剪切分离的发生,局部的低压区驱使周围空气流入近尾流区,加之与底部区域流动的相互作用,从而形成螺旋流动[12,30].在这个过程中,流体微团因剪切作用而发生的变形是显著的,同时边界层流动的分离也使得较高的涡量在尾端附近进行扩散,这些为较强涡旋的形成和脱落提供了重要条件,正如图9所示,流向涡旋的形成是局部流动中重要的物理过程,以至于涡边界($\varOmega= 0.52$)未出现,即变形与涡量是相当的.不过,较大的剪切应变往往因为黏性效应而具有高耗散性,相应的涡量会有较为明显的衰减,并且在下游位置,对已形成的涡旋而言,图8中更大$\varOmega $ 值表明,较弱的涡旋是尾涡结构的主要部分,它们的涡量超过流体微团的变形,并且如图9(b)~图9(c)所示,涡旋涡量的占比增大,意味着剪切的影响作用在下降,虽然涡核的强度明显减小,但此时的耗散较低且旋转运动具有主导影响,这对尾涡结构能够处于相对稳定的状态是很重要的. 显示原图|下载原图ZIP|生成PPT 图9不同流向位置处(x = 0.5 H, 2H, 4 H)的流向涡核(时均$Q$和 $\varOmega $等值线分别对应$0.5 Q_{max}$和0.52) -->Fig. 9Streamwise vortex cores at the varied streamwise positions x = 0.5 H, $2 H, 4 H (Time-averaged $Q$ and $\varOmega $ contour lines represent $0.5 Q_{max}$ and 0.52, respectively) -->
Tennekes 和Lumley[35]认为,涡旋须要通过剪切的旋转和拉伸来维持自身的能量,研究表明[32,34],靠近尾车处的雷诺应力和湍流能量产生速率最为突出,但同时下降得也十分迅速,这一定程度上反映出局部剪切层的强度和耗散性均是很显著的.然而,在离开尾车的下游,依然会有较低水平的雷诺应力以及湍流产生过程[32,34],也就是说,还有弱剪切对涡旋获取平均流能量发挥着作用,鉴于此,当低涡量涡核较靠近地面,并且旋转运动发挥主导作用时,涡旋与地面间的相互作用[36-37]被认为是维持局部流动剪切的重要机制. 图10~图12分别给出了不同流向位置处湍流产生的$y$-$z$平面时均分布(其中,$P_{xy}$和$P_{xz}$分别表征涡旋在$x$-$y$平面和$x$-$z$平面平均剪切($S_{xy}$和$S_{xz}$)的作用下对湍流产生的贡献,$P$代表局部总的湍流产生).如图10所示,在靠近尾车的位置($x = 0.5H$),$P_{xy}$和$P_{xz}$分别对局部的湍流产生过程有较大的贡献,而相应的分离剪切层应该与模型底部和侧表面上发展起来的边界层流动有很大的关联性.而通过图11~图12可观察到,随着流向涡的移动,湍流产生较为显著的位置展向外移,同时,$P$曲线的峰值明显下移,并且局部的涡旋主要依赖于$x$-$y$平面剪切在产生湍流能量方面作出突出的贡献.基于此,我们认为,通过流向涡与地面之间的相互作用,局部速度场以及应力场受到扰动,虽然涡旋强度会不可避免地逐渐衰减,但重要的是,剪切能够得以维系,使局部的涡旋继续从平均流动中获取能量,这样有助于实现涡旋的自维持,从而尾涡结构能够处于相对稳定的状态. 显示原图|下载原图ZIP|生成PPT 图10在流向位置$x = 0.5 H$,$P$, $P_{xz}$和$P_{xy}$的y-z平面时均分布 -->Fig. 10Time-averaged $y$-$z$ plane profiles of the $P$, $P_{xz}$, and $P_{xy}$ at $x = 0.5 H$ -->
显示原图|下载原图ZIP|生成PPT 图11在流向位置$x = 2 H$,$P$, $P_{xz}$和$P_{xy}$的$y$-$z$平面时均分布 -->Fig. 11Time-averaged $y$-$z$ plane profiles of the $P$, $P_{xz}$, and $P_{xy}$ at $x = 2 H$ -->
显示原图|下载原图ZIP|生成PPT 图12在流向位置$x = 4 H$,$P$, $P_{xz}$和$P_{xy}$的$y$-$z$平面时均分布 -->Fig. 12Time-averaged $y$-$z$ plane profiles of the $P$, $P_{xz}$, and $P_{xy}$ at $x = 4 H$ -->
4 结 论
本文通过IDDES对基于高速列车气动外形的简化缩尺模型的绕流流场进行模拟,以获取尾流流动的数值结果,进而利用涡旋识别方法,对高速列车尾涡结构进行研究. 可以观察到,在靠近尾车的局部流动中分布有高涡量的尾流涡旋,而随着尾涡结构向下游发展,其中更多的是低涡量涡旋,它们分布在较大的近尾流区域中,反映出尾涡结构处于相对稳定状态. 结合无量纲参数 $\varOmega $ 的物理意义,我们认为:边界层分离是尾端附近复杂流动的关键机制,即一方面,与之相关的剪切变形以及高涡量的扩散对具有较大强度的尾涡的形成起到了重要作用,也正因如此,局部尾流具有很显著的湍流特性;不过,对于低涡量涡旋而言,由于强剪切的高耗散,总涡量中的涡旋涡量是主要部分,相对于变形,流体微团的旋转运动是主导的,此时,耗散较为缓慢,并且重要的是,在流向涡核接近地面的情况下,涡旋与地面间的相互作用成为涡旋自维持的重要机制,使尾涡结构能够在向下游发展的过程中处于相对稳定的状态. The authors have declared that no competing interests exist.
(YaoShuanbao, GuoDidong, YangGuowei. Aerodynamic optimization of high speed train based on RBF mesh deformation ., 2013, 45(6): 982-986 (in Chinese)) [本文引用: 1]
(AnYi, MoHuangrui, LiuQingquan. Study on the influence of the nose slenderness ratio of high-speed train on the aerodynamic noise ., 2017, 49(5): 985-996 (in Chinese)) [本文引用: 2]
[4]
YuHuahua, LiJiachun, ZhangHuiqin. On aerodynamic noises radiated by the pantograph system of high-speed trains ., 2013, 29(3): 399-410 [本文引用: 1]
(YangGuowei, WeiYujie, ZhaoGuilin, et al. Current research progress in the mechanics of high speed rails ., 2015, 45: 217-460 (in Chinese)) [本文引用: 2]
[7]
BakerCJ, DalleySJ, JohnsonT, et al. The slipstream and wake of a high-speed train ., 2001, 215(2): 83-99. [本文引用: 1]
[8]
BakerCJ. The flow around high speed trains ., 2010, 98(6-7): 277-298 [本文引用: 3]
[9]
BakerCJ, QuinnA, SimaM, et al. Full-scale measurement and analysis of train slipstreams and wakes. Part 1: Ensemble averages ., 2014, 228(5): 451-467 [本文引用: 1]
[10]
Schulte-WerningB, HeineB, MatschkeC. Slipstream development and wake flow characteristics of modern high-speed trains ., 2001, 81(S3): 789-790 [本文引用: 1]
[11]
Schulte-WerningB, HeineB, MatschkeC. Unsteady wake flow characteristics of high-speed trains ., 2003, 2(1): 332-333 [本文引用: 1]
[12]
BellJR, BurtonD, ThompsonM, et al. Wind tunnel analysis of the slipstream and wake of a high-speed train ., 2014, 134: 122-138 [本文引用: 4]
[13]
BellJR, BurtonD, ThompsonMC, et al. Moving model analysis of the slipstream and wake of a high-speed train ., 2015, 136: 127-137 [本文引用: 2]
[14]
BellJR, BurtonD, ThompsonMC, et al. Flow topology and unsteady features of the wake of a generic high-speed train ., 2016, 61: 168-183 [本文引用: 2]
[15]
BellJR, BurtonD, ThompsonMC, et al. Dynamics of trailing vortices in the wake of a generic high-speed train ., 2016, 65: 238-256 [本文引用: 2]
[16]
MuldTW. Analysis of Flow Structures in Wake Flows for Train Aerodynamics. Stockholm, , 2010 [本文引用: 3]
[17]
MuldTW, EfraimssonG, DanSH. Flow structures around a high-speed train extracted using proper orthogonal decomposition and dynamic mode decomposition ., 2012, 57(4): 87-97 [本文引用: 3]
[18]
MuldTW, EfraimssonG, HenningsonDS. Wake characteristics of high-speed trains with different lengths ., 2014, 228(4): 333-342 [本文引用: 3]
[19]
XiaC, WangH, ShanX, et al. Effects of ground configurations on the slipstream and near wake of a high-speed train ., 2017, 168:177-189 [本文引用: 1]
(TianHongqi. Development of research on aerodynamics of high-speed rails in China . , 2015(4): 30-41 (in Chinese)) [本文引用: 1]
[21]
MorelT. Effect of base slant on flow in the near wake of an axisymmetric cylinder ., 1980, 31: 132-147 [本文引用: 1]
[22]
PauleyWR, EatonJK. Experimental study of the development of longitudinal vortex pairs embedded in a turbulent boundary layer ., 1988, 26(7): 816-823 [本文引用: 1]
[23]
RoweA, FryALA, MotallebiF. Influence of boundary-layer thickness on base pressure and vortex shedding frequency ., 2001, 39(4): 754-756 [本文引用: 1]
[24]
KüchemannD. Report on the I.U.T.A.M. symposium on concentrated vortex motions in fluids ., 1965, 21(1): 1-20 [本文引用: 1]
[25]
LiuCQ, WangYQ, YangY, et al. New omega vortex identification method ., 2016, 59(8): 684711 [本文引用: 5]
[26]
HuntJCR, WrayAA, MoinP. Eddies, streams, and convergence zones in turbulent flows// , 1988: 193-208 [本文引用: 2]
[27]
ChongMS, PerryAE, CantwellBJ. A general classification of three-dimensional flow fields ., 1990, 2(5): 765-777 [本文引用: 1]
[28]
JeongJ, HussainF. On the identification of a vortex ., 1995, 285(4): 69-94 [本文引用: 2]
[29]
YaoSB, SunZX, GuoDL, et al. Numerical study on wake characteristics of high-speed trains ., 2013, 29(6): 811-822 [本文引用: 1]
[30]
HemidaH, BakerC, GaoG. The calculation of train slipstreams using large-eddy simulation ., 2014, 228(1): 25-36 [本文引用: 1]
[31]
HuangS, HemidaH, YangM. Numerical calculation of the slipstream generated by a CRH2 high-speed train . , 2016, 230(1): 103-116 [本文引用: 1]
(PanYongchen, YaoJianwei, LiChangfeng. Numerical simulation and dynamics discussion of turbulent eddies in the near wake of a high speed train . , 2017(2): 351-358 (in Chinese)) [本文引用: 4]
[33]
RobinsonSK. Coherent motions in the turbulent boundary layer ., 1991, 23: 601-639 [本文引用: 1]
(PanYongchen, YaoJianwei, LiangCe, et al. Analysis on turbulence characteristics of vortex structures in the near wake of a high-speed train ., 2017, 38(2): 83-88 (in Chinese)) [本文引用: 3]
[35]
TennekesH, LumleyJ. Cambridge: MIT Press, 1972 [本文引用: 1]
[36]
CorjonA, PoinsotT. Behavior of wake vortices near ground ., 1997, 35(5): 849-855 [本文引用: 1]
[37]
PerkinsHJ. The formation of streamwise vorticity in turbulent flow ., 1970, 44(4): 721-740 [本文引用: 1]