【作者】 肖红林；

【导师】 周恒；

【作者基本信息】 天津大学 ， 流体力学， 2004， 硕士

【摘要】 目前，工程中大部分的湍流计算依靠湍流模式理论，但常用的模式理论缺少通用性；而直接数值模拟又耗用大量机时，故需要一种运算时间较少、通用性较强的方法。大涡模拟正是这样一种方法，它首先由气象学家Samgorinsky（1963）提出。其基本思想是把包括脉动在内的湍流瞬时流动通过某种滤波的方法分解为大尺度运动和小尺度运动两部分，大尺度直接求解方程，小尺度建立模型来模拟。本文通过大涡数值模拟的方法，模拟了不同雷诺数下槽道湍流流场，通过分析不同雷诺数下的数和数，研究了壁面率问题。研究了不同雷诺数下的雷诺应力分布，发现在半槽宽坐标下，随着雷诺数的增加，雷诺应力峰值向壁面方向移动；而在壁面坐标下，随着雷诺数的增加，雷诺应力峰值向远离壁面方向移动。本文还对不同雷诺数下均方根速度的分布进行了研究；给出了不同雷诺数的平均速度剖面。本文用了较少的网格点进行数值模拟，得到了较好的结果，这对大涡模拟在工程实际中的应用具有重要意义。更多还原

【Abstract】 Most practical computation of turbulent flows relies on turbulence modeling. However, there is no universally valid model for turbulence, while direct numerical simulation needs vast resources of computer which makes it unpractical. Thus, we need a method which needs only a reasonable computer resources and is sufficiently universal. Large eddy simulation is just such a method, which was first proposed by meteorologist Smagorinsky in 1963. Its basic idea is that the instantaneous turbulent quantities, including pulsating quantities, can be decomposed into two parts, namely, large scale motion and small scale motion, by some kind of filtering method. The large scale motion should be computed directly, while the small scale motion should be modeled. In this thesis, by means of LES, turbulent channel flow were computed for several different Reynolds numbers. The wall law has been examined by analyzing the dependence of and on Reynolds numbers. The Reynolds stress distribution for different Reynolds number were analyzed, and we found that the position of the peak of the Reynolds stress was closer to the wall when the Reynolds number increased, when measured by channel width. On the other hand, the position of the peak of Reynolds stress moves away from the wall when measured by wall units. The rms of fluctuating velocity, as well as the mean velocity profiles, for different Reynolds number has also been studied.Our LES computation used relatively few grid points while the results were reasonably good, which is important for the practical of LES computation for engineering purposes.更多还原