VORTEX IDENTIFICATION TECHNOLOGY AND ITS APPLICATION IN THE WAKE FIELD OF MARINE PROPELLER
-
摘要: 涡识别是很重要的流体问题, 为了在船用螺旋桨伴流场中寻找一种合理的涡识别方法, 本文结合实践, 研究了六种涡识别技术理论, 其中使用Burgers涡流和Lamb-Oseen涡流作了必要的解释, 讨论了各种识别方法的优缺点. 局部低压标准比较直观, 但深究其黏性和非定常影响后, 明显不足; 迹线或流线显然不能满足伽利略不变性, 会使辨别变得混乱; 涡度大小需要规定其阈值, 具有一定不确定性, 且也会识别不是涡的涡片; 速度梯度张量的复特征值也会有识别不出的区域; 速度梯度张量的第二不变量标准和对称张量的第二特征值标准能更好地识别涡核, 这两种标准有时等效. 螺旋桨伴流场的数值模拟是在开源软件OpenFOAM平台上实现的, 湍流大涡模型由一种局部动态方程建模, 此模型优于动态Smagorinsky模型. 最终的结果显示: 对于船用螺旋桨伴流场中的涡, 采用速度梯度张量的第二不变量和对称张量的第二特征值的结果基本一致, 而最小压力标准、流线或迹线标准、涡度值标准和速度张量的复特征值标准都存在一定的缺陷, 不适用于船用螺旋桨伴流场中的涡识别.Abstract: Vortex identification is a very important problem of fluid and flow, in order to find a reasonable method of vortex identification in the wake of marine propeller, this paper studies the theory of six kinds of vortex recognition technology combined with practice, in which analytic solutions of both Burgers vortex and Lamb-Oseen vortex are also used for necessary explanation. The advantages and disadvantages of various vortex identification methods are discussed in detail at the angle of theory and application. The local low-pressure criterion is intuitive, but it is obviously insufficient after considering viscous and unsteady effects. The path line or streamline criterion obviously cannot satisfy Galileo invariance, which will cause confusion in vortex identification. The magnitude of vorticity criterion needs to specify its threshold value, which has certain uncertainty, and can also incorrectly identify vortex sheets that are not vortices. The complex eigenvalue of the velocity gradient tensor will also have an unrecognized region. The second invariant criterion of the velocity gradient tensor and the second eigenvalue criterion of the symmetric tensor can better identify the vortex core, and these two criteria are sometimes equivalent. The numerical simulation of propeller wake is implemented on the open source software OpenFOAM platform. The large eddy model is modeled by a local dynamic equation, which is better than the dynamic Smagorinsky model to a certain extent. The results of numerical experiment show that, for the vortex identification in the marine propeller wake, the second invariant criterion of the velocity gradient tensor is consistent with the second eigenvalue criterion of the symmetric tensor. However, the local minimum pressure criterion, streamline or path line criterion, vorticity magnitude criterion and complex eigenvalue criterion of velocity gradient tensor have some defects, which are not suitable for vortex identification in the wake of marine propeller.
-
Key words:
- vortex identification /
- marine propeller /
- wake field /
- large eddy simulation
-
图 5 相对涡度和压力分布(实线为涡度, 虚线为压力, 黑虚线为
$ {\omega }_{0} = 1 $ , 向下橙紫绿黄蓝$ {\omega }_{0} $ 分别为$ 2,3, \cdots ,6 $ )Figure 5. Relative vorticity and pressure distribution (solid line is vorticity; dotted line is pressure; black dotted line is
$ {\omega }_{0} = 1 $ ; downward orange-purple-green-yellow-blue lines are$ {\omega }_{0} = 2,3, \cdots, 6 $ respectively)表 1 涡识别方法
Table 1. Methods of vortex identification
No. Criterions of identifying vortex 1 local minimum pressure 2 pathline or streamline 3 vorticity magnitude 4 complex eigenvalue of tensor $ \boldsymbol{T} $ 5 second invariant $ {\boldsymbol{Q}} $ or kinematic vorticity number $ {N}_{k} $ 6 eigenvalue of sym tensor $\left({\boldsymbol{S} }^{2} + {\boldsymbol{\varOmega } }^{2}\right)$ 表 2 计算区不同域网格尺度
Table 2. Overall cell scales in different regions of computational domain
Cell scales for about total 8.5 million cells Region 0 (background grid) $ 0.125 D $ Region 1 (dashed line) $ 0.015\;6 D $ Region 2 (most refinement) $ 0.007\;81 D $ 表 3 基于负
$ {\lambda }_{2} $ 和正$ Q $ 标准识别涡核对比Table 3. Comparison of vortex core identifing results based on negative
$ {\lambda }_{2} $ and positive$ Q $ criteria$ {\lambda }_{1} $ $ {\lambda }_{2} $ $ {\lambda }_{3} $ $\displaystyle \sum {\lambda }_{\mathrm{i} }$ $ -{\lambda }_{2} $ $ + Q $ $ + $ $ - $ $ - $ $ - $ yes yes $ + $ $ - $ $ - $ $ + $ yes no $ + $ $ + $ $ - $ $ - $ no yes $ + $ $ + $ $ + $ $ + $ no no -
[1] Tennekes H, Lumley JL. A First Course in Turbulence. Cambridge, Massachusetts, and London, England: MIT Press, 1972: 59-103 [2] Hussain F, Melander MV. Studies in Turbulence: Understanding Turbulence via Vortex Dynamics. New York: Springer-Verlag Inc., 1992: 157-178 [3] Lamb SH. Hydrodynamics, Sixth edition. The United States of America: Syndics of the Cambridge University Press, 1975: 202-249 [4] Lugt HJ. Recent Developments in Theoretical and Experimental Fluid Mechanics: The Dilemma of Defining a Vortex. Berlin Heidelberg: Springer-Verlag, 1979: 309-321 [5] Chong MS, Perry AE, Cantwell BJ. A general classification of three-dimensional flow fields. Physics of Fluids A: Fluid Dynamics, 1990, 2(5): 765-777 doi: 10.1063/1.857730 [6] Hunt JCR, Wray AA, Moin P. Eddies, streams, and convergence zones in turbulent flows//Proceeding of the Summer Program, Center for Turbulence Research, Stanford University, 1988: 193-208 [7] Jeong J, Hussain F. On the identification of a vortex. Journal of Fluid Mechanics, 1995, 285: 69-94 doi: 10.1017/S0022112095000462 [8] Cantwell BJ. Organized motion in turbulent flow. Annual Review of Fluid Mechanics, 1981, 13: 457-515 doi: 10.1146/annurev.fl.13.010181.002325 [9] Lumley JL. Transition and Turbulence: Coherent Structures in Turbulence. Academic Press Inc., 1981: 215-242 [10] Hussain AKMF. Coherent structures and turbulence. Journal of Fluid Mechanics, 1986, 173: 303-356 doi: 10.1017/S0022112086001192 [11] Mumford JC. The structure of the large eddies in fully developed turbulent shear flow. Part 1: The plane jet. Journal of Fluid Mechanics, 1982, 118: 241-268 doi: 10.1017/S0022112082001062 [12] Mumford JC. The structure of the large eddies in fully developed turbulent shear flow. Part 2: The plane wake. Journal of Fluid Mechanics, 1983, 137: 447-456 doi: 10.1017/S0022112083002505 [13] Blackwelder R. On the role of phase information in conditional sampling. Physics of Fluids, 1977, 20(10): S232-242 doi: 10.1063/1.861736 [14] Fiedler HE, Mensing P. The plane turbulent shear layer with periodic excitation. Journal of Fluid Mechanics, 1985, 150: 281-309 doi: 10.1017/S0022112085000131 [15] Ferré JA, Giralt F. Pattern-recognition analysis of the velocity field in plane turbulent wakes. Journal of Fluid Mechanics, 1989, 198: 27-64 doi: 10.1017/S0022112089000029 [16] Kim J. Turbulence structures associated with the bursting event//NASA Ames Research Center, Moffett Field, California 94035, 1984, NASA Technical Memorandum 85944 [17] Hussain AKMF, Zaman KBMQ. Vortex pairing in a circular jet under controlled excitation. Part 2: Coherent structure dynamics. Journal of Fluid Mechanics, 1980, 101: 493-544 doi: 10.1017/S0022112080001772 [18] Cantwell B, Coles D. An experimental study of entrainment and transport in the turbulent near wake of a circular cylinder. Journal of Fluid Mechanics, 1983, 136: 321-374 doi: 10.1017/S0022112083002189 [19] Tso J, Hussain F. Organized motions in a fully developed turbulent axisymmetric jet. Journal of Fluid Mechanics, 1989, 203: 425-448 doi: 10.1017/S0022112089001539 [20] Hussain AKMF, Hayakawa M. Eduction of large-scale organized structures in a turbulent plane wake. Journal of Fluid Mechanics, 1987, 180: 193-229 doi: 10.1017/S0022112087001782 [21] Robinson SK. The kinetics of turbulent boundary layer structure. [PhD Thesis]. Stanford University, 1991 [22] 杨越. 湍流中的涡识别与涡动力学//第十届全国流体力学青年研讨会, 2017Yang Yue. Vortex identification and vortex dynamics in turbulence//The 10th National Youth Symposium on Fluid Mechanics, 2017 (in Chinese) [23] 王义乾, 桂南. 第三代涡识别方法及其应用综述. 水动力学研究与进展: A辑, 2019, 34(4): 413-429 (Wang Yiqian, Gui Nan. A review of the third-generation vortex identification method and its applications. Chinese Journal of Hydrodynamics:Series A, 2019, 34(4): 413-429 (in Chinese) [24] 刘超群. Liutex-涡定义和第三代涡识别方法. 空气动力学学报, 2020, 38(3): 405-431,478 (Liu Chaoqun. Liutex-third generation of vortex definition and identification methods. Acta Aerodynamics Sinica, 2020, 38(3): 405-431,478 (in Chinese) [25] Pang C, Yang H, Gao ZH, et al. Enhanced adaptive mesh refinement method using advanced vortex identification sensors in wake flow. Aerospace Science and Technology, 2021, 115: 106796 [26] Bai XR, Cheng HY, Bin J. Liutex and Third Generation of Vortex Definition and Identification: On the Comparison of Liutex Method with Other Vortex Identification Methods in a Confined Tip-Leakage Cavitating Flow. Cham Springer, 2021: 139-155 [27] Zhang F, Appiah D, Chen K, et al. Investigation on the flow behavior of side channel pumps based on vortex identification. Chinese Journal of Mechanical Engineering, 2021, 34(6): 1-18 [28] Zhang YH, Hu XJ, Lan W, et al. Application of omega vortex identification method in cavity buffeting noise. Journal of Hydrodynamics, 2021, 33(2): 259-270 doi: 10.1007/s42241-021-0034-8 [29] Wang YQ, Gao YS, Xu HY, et al. Liutex theoretical system and six core elements of vortex identification. Journal of Hydrodynamics, 2020, 32(2): 197-211 doi: 10.1007/s42241-020-0018-0 [30] Asmaee Z, Delda S. Introducing vortices in the continuum using direct and indirect methods. Physical Review D, 2022, 105: 096020 [31] Fu SM, Zhang JP, Tang H, et al. A new mesoscale-vortex identification metric: restricted vorticity and its application. Environmental Research Letters, 2020, 15(2020): 124053 [32] Yusuke N, Takuji N, Takenori H, et al. Identification of the vortex around a vehicle by considering the pressure minimum. Journal of Visualization, 2020, 23(5): 793-804 doi: 10.1007/s12650-020-00665-8 [33] Ye ZX, Chen Q, Li BH, et al. Flow structure segmentation for vortex identification using butterfly convolutional neural networks. International Journal of Modern Physics B, 2020, 34(14/16): 2040121 [34] Yang K, Wu SQ, Zhang H, et al. Lagrangian-averaged vorticity deviation of spiraling blood flow in the heart during isovolumic contraction and ejection phases. Medical & Biological Engineering & Computing, 2021, 59(7-8): 1417-1430 [35] Yu YF, Shrestha P, Alvarez O, et al. Liutex and Third Generation of Vortex Definition and Identification: Incorrectness of the Second-Generation Vortex Identification Method and Introduction to Liutex. Cham Springer, 2021: 37-56 [36] Kim WW, Menon S. A new dynamic one-equation subgrid-scale model for large eddy simulation//33rd Aerospace Sciences Meeting and Exhibit, 1995: 1-9 [37] Sagaut P. Large Eddy Simulation for Incompressible Flows. Second Edition. Berlin Heidelberg Springer-Verlag: GmbH, 2006: 39-73 [38] Bridges DH. A detailed study of the flowfield of a submarine propeller during a crashback maneuver. [PhD Thesis]. Department of Aerospace Engineering, Mississippi State University, 2004 [39] Moffatt HK. Viscous and resistive eddies near a sharp corner. Journal of Fluid Mechanics, 1964, 18: 1-18 doi: 10.1017/S0022112064000015 [40] Panton RL. Incompressible Flow. 2nd edition. New York: Wiley, 1996: 837 [41] Melander MV, Hussain F. Cut-and-connect of two antiparallel vortex tubes: a new cascade mechanism//Seventh International Symposium on Turbulent Shear Flows 7. Springer-Verlag Berlin Heidelberg, 1991: 9-16 [42] Metcalfe RW. Coherent structures in a turbulent mixing layer: A comparison between numerical simulations and experiments. Turbulent Shear Flows, 1985, 5: 110-123 [43] Bisset DK, Antonia RA, Browne LWB. Spatial organization of large structures in the turbulent far wake of a cylinder. Journal of Fluid Mechanics, 1990, 218: 439-461 doi: 10.1017/S0022112090001069 [44] Jimenez J, Moin P, Moser R, et al. Ejection mechanisms in the sublayer of a turbulent channel. Physics of Fluids, 1988, 31(6): 1311-1313 doi: 10.1063/1.866721 [45] Truesdell CA. The kinematics of vorticity. Bulletin of the American Mathematical Society, 1949, 55(3): 296-296 [46] Melander MV, Hussain F. Polarized vorticity dynamics on a vortex column. Physics of Fluids A: Fluid Dynamics, 1993, 5(8): 1992-2003 doi: 10.1063/1.858826 [47] Courant R, Hilbert D. Methods of mathematical physics, Vol. I. Physics Today, 1954, 7(5): 17 [48] Crowdy DG. A note on the linear stability of burgers vortex. Studies in Applied Mathematics, 1996, 100: 107-126 [49] Fabre D, Sipp D, Jacquin L. Kelvin waves and the singular modes of the Lamb-Oseen vortex. Journal of Fluid Mechanics, 2006, 551: 235-274 doi: 10.1017/S0022112005008463 [50] Batchelor GK. An Introduction to Fluid Dynamics. Cambridge University Press, 2002: 266-273 -