簡易檢索 / 詳目顯示
| 研究生： |
姚立謙 Yao, Li-Chang |
|---|---|
| 論文名稱： |
大尺度渦流模擬法於圓柱紊態尾流場之分析 Analysis of Flow over a Cylinder by Large Eddy Simulation (LES) |
| 指導教授： |
張克勤
Chang, Keh-Chin |
| 學位類別： |
碩士 Master |
| 系所名稱： |
工學院 - 航空太空工程學系 Department of Aeronautics & Astronautics |
| 論文出版年： | 2016 |
| 畢業學年度： | 104 |
| 語文別： | 中文 |
| 論文頁數： | 99 |
| 中文關鍵詞： | 大尺度渦流模擬法 、圓柱尾流 、進口條件 、實驗儀器比較 、次格點模型 |
| 外文關鍵詞： | large eddy simulation, wake of a circular cylinder, inlet conditions, comparison of different anemometry, sub-grid model |
| 相關次數： | 點閱：59 下載：8 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
本計算研究以大尺度渦流模擬法(LES)計算不同雷諾數 Re =9959 與Re = 3856於尾流之平均速度和平方均根擾動速度，而雷諾數之定義使用入口速度 與圓柱之直徑D 來定義，其結果並與石昌隆(2015)之實驗值做比較。
計算以506644格點數作為基礎樣本，先使用次格點模型dynamic Smagorinsky model；而進口條件為無側流與橫流向平均速度之實驗值，其方法保持其紊流特性，去除實驗值較易造成誤差因素，而因此比較之進口條件為使用固定進口速度(fixed profile)與完全實驗值之進口。其三者綜合觀察下，最符合實驗值者為無側、橫流平均速度之實驗值進口速度。網格設置方面，與全域加密之983098格點數、尾流區塊部分加密做比較，得到於尾流各截面平均速度與平方均根擾動速度三者結果差異不大，但可以從迴流長度部份發現，網格之差異將影響迴流長度大小。此外本研究也針對不同次格點模型做比較，另一組參照組使用dynamic kinetic energy sub-grid scale model，發現尾流截面並無明顯差異，其原因推估為流場基本特色單純；但是從頻譜上可以推估，若於不同流場，使用不同次格點模型將有明顯之影響。
This study is aimed to evaluate the significance of inlet conditions on the prediction accuracy for large eddy simulation and compare results with the experimental measurement at Reynolds number 3856 and 9959, which are in the range of sub-critical regime of wake flow. The control factors of computation include (1) grid number of 506644,983098, and refined mesh in near wake subregion, (2) inlet conditions given by assumed and experimental measured velocity profiles, and (3) difference between the sub-grid models of dynamic Smagorinsky model (DSM) and dynamic kinetic energy model (DKEM).
The effect of the grid size on LES prediction is examined with the power spectral density of velocity at specific points. A small enough grid size for LES would show a slope of -5/3 at the range of frequencies from 1000 to 10000 Hz in the power spectral density of velocity. According to the conditions of power spectral density, all the samples in the test of grid are found to fall within Taylor’s micro-scale. Thus, the resolutions of large-scale eddies are reliable. From the velocity distributions of simulation results, it is found that there is no obvious difference in the test of various grid sizes. However, distinct grid size would influence the recirculation bubble length slightly in the simulation test. Among inlet conditions, it is found that experimental measured velocity profiles with zero value of the transverse and span-wise mean velocities but keeping their fluctuating velocities are the closest to the experiment results at the five selected stream-wise sections, since there are only upper half measured inlet velocity profiles from the measurements. In the comparison of different sub-grid model, it is discovered that these two simulated results are very close each other. Since the regime of wake past a circular cylinder is not very complex, the influence of the different sub-grid models is not apparent in this simulation research.
[1] Parnaudeau et al., “Experimental and Numerical Studies of the Flow over a Circular Cylinder at Reynolds Number 3900.”,2008
[2] Lysenko, Ertesvåg, and Rian, “Large-Eddy Simulation of the Flow over a Circular Cylinder at Reynolds Number 3900 Using the Openfoam Toolbox.”,2012
[3] Dong et al., “A Combined Direct Numerical Simulation–particle Image Velocimetry Study of the Turbulent near Wake.”,2006
[4] Kornhaas, Sternel, and Sch, Influence of Time Step Size and Convergence Criteria on Large Eddy Simulations with Implicit Time Discretization.,2008
[5] Breuer, “Large Eddy Simulation of the Subcritical Flow Past a Circular Cylinder : Numerical and Modeling Aspects.”,1998
[6] Leonard,“The Ultimate conservative difference scheme applied to unsteady one dimensional advection.pdf.”,1991
[7] Franke, “Large Eddy Simulation of the Flow Past a Circular Cylinder at ReD=3900.”,2002
[8] 劉育政, “3D Large Eddy Simulation of Mixing Layer Flow Field.”,2009
[9] Montorfano, Piscaglia, and Ferrari, “Inlet Boundary Conditions for Incompressible LES,: A Comparative Study.”
[10] 陳宇杰, “Large Eddy Simulation of Turbulent Backward-Facing Step Flow.”,2013
[11] Mathey et al., “Assessment of the Vortex Method for Large Eddy Simulation Inlet Conditions.”,2006
[12] 石昌隆, “Investigation of Velocity Measurement Technology in Turbulent Wake Over the Circular Cylinider.”,2015
[13] M. Germano, U. Piomeli, P. Moin, and W. Cabot, “A Dynamic Subgrid-Scale Eddy Viscosity Model.”,1991
[14] Kim and Menon, “Application of the Localized Dynamic Subgrid-Scale Model to Turbulent Wall-Bounded Flows.”,1997
[15] ANSYS Inc., ANSYS FLUENT 15.0.0 User’s Guide.,2013
[16] Pope, “Ten Questions Concerning the Large-Eddy Simulation of Turbulent Flows.”,2004
[17] Kim and Menon, “Application of the Localized Dynamic Subgrid-Scale Model to Turbulent Wall-Bounded Flows.”,1997
[18] Sakamoto, H. Haniu, Sakamoto, and Haniu, “A Study on Wortex Shedding From Spheres in a Uniform Flow.”,1990
[19] Yoshie, R., Tanaka,“Technique For Simultaneously Measuring Fluctuating Velocity.”,2007
[20] Frohlich et al., “Large Eddy Simulation of Flow around Circular Cylinders on Structured and Unstructured Grids, II.”,2001
校內：立即公開校外：不公開