圓柱體在不同雷諾數下之表面風壓特性數值模擬

Numerical Simulation of surface pressure on circular cylinders of different Reynolds numbers

豐喬(Chiao Feng)

指導教授 ： 鄭啟明

淡江大學/工學院/土木工程學系碩士班/碩士(2011年)

https://doi.org/10.6846/TKU.2011.00929

摘要

當鈍體受到風力作用時，由於物體的形狀或是雷諾數的大小，流場會產生不同位置的分離及尾跡產生，而本文將以數值模擬方法來分析通過圓柱體的複雜流場。探討的雷諾數區間則是於次臨界流及超臨界流下之流場，即雷諾數為6.1x10^4和10^6。
本文採用了不同的紊流模式，包含RNG k-epsilon、LES、Laminar model來進行模擬，並嘗試不同的邊界尺寸來改善其邊界對於流場模擬的影響。文中亦和實驗值與其他數值模擬的參考文獻進行比較，其平均風壓係數的分佈及阻力係數、升力係數和史托赫的計算值。
研究結果顯示，當流場為次臨界流時，流體剛好在層流與紊流轉換的一個狀態，所以使用Laminar和適用高雷諾數的RNG k-epsilon模式都無法得到好結果，而使用LES模式在平均風壓分佈及風力係數的計算上，都能和實驗有相近的結果。於此雷諾數下建議可使用User Define Functions (UDF)來進行適合的公式來修改計算式，避免結果偏差於軟體之設定，或是要求的細密網格超過可接受的時間成本。
另外在超臨界流下，使用二維的RNG k-epsilon已可以得到和實驗值十分吻合的結果。而若要清楚的將流場可視化，則須配合適當的計算域大小及優良網格繪製，避免邊界過於狹隘造成流場壓縮。由於邊界的影響在風壓分佈上並不明顯，建議在進行後處理計算前可以先於繪圖軟體檢視流場的合理性，可節省試誤的時間。

關鍵字

數值模擬 ； 雷諾數 ； 次臨界流 ； 超臨界流 ； 表面風壓

並列摘要

When the wind passed through a curve shaped bluff body, at different Reynolds number, separation occurs at different location. In this study, numerical method is used to simulate the flow field around a circular cylinder at both subcritical and supercritical Reynolds number region, 6.1x10^4 and 1.0x10^6, respectively. Besides turbulence models such as RNG k-epsilon and LES, Laminar model were also used in this study. Several combinations of numerical simulation domain and grid structures were studied to ensure better numerical result. The simulated mean pressure distribution, drag coefficient, lift coefficient and Strouhal number were compared with data reported in literatures.

When the flow is in the sub-critical region, the thin boundary layer on cylinder surface is laminar and later transits to turbulent after separation. Therefore, the Laminar model was used. The Laminar model produces good results up to the point of separation but cannot correctly simulate the wake flow. Comparing to the experimental data, LES model produces better results. In the future, the User Define Functions (UDF) mode should be activated to introduce appropriate modification of the Fluent’s general purposed parameter setting When the flow is the super-critical region, two-dimensional RNG k-epsilon model can produce results in good agreement with experimental data. It was found that, appropriate computational domain and sufficient fine grid are needed in order to obtain good pictures of the flow field visualization.

並列關鍵字

Numerical Simulation ； Reynolds numbers ； Sub-critical ； Super-critical ； surface pressure

參考文獻

[1] S. Murakami, A. Mochida “On turbulent vortex shedding flow past 2D square cylinder predicted by CFD” Journal of Wind Engineering and Industrial Aerodynamics 54/55 (1995) 191-211.

[2] D. Bouris, G. Bergeles “2D LES of vortex shedding from a square cylinder” Journal of Wind Engineering and Industrial Aerodynamics 80 (1999) 31-46.

[3] U. Schumann “ Subgrid scale model for pnite difference simulations of turbulent flows in plane channels and annuli ” J. Comp. Phys. 18 (1975) 376-404.

[4] Yi Yang a, MingGu ,, SuqinChen , XinyangJin “New inflow boundary conditions for modelling the neutral equilibrium atmospheric boundary layer in computational wind engineering” Journal of Wind Engineering and Industrial Aerodynamics 97 (2009) 88–95

[5] Richards, P.J., Quinn, A.D., “A 6m cube in an atmosphere boundary layer flow, part 2. Computational solutions.” Wind and Structures 5, 177–192.