湍流大涡模拟的时空关联方法

doi:10.1360/SSPMA2015-00446

IMECH-IR > 非线性力学国家重点实验室

	湍流大涡模拟的时空关联方法
其他题名	Space-time correlation method for large-eddy simulation of turbulent flows
	晋国栋; 陈进财; 时北极; 何国威
发表期刊	中国科学: 物理学力学天文学
	2015-11-05
卷号	45 期号:12 页码:124702
摘要	本文介绍了湍流大涡模拟的时空关联方法, 它用于检验和发展大涡模拟的亚格子尺度模型, 其目的是提高大涡模拟方法对于欧拉和拉格朗日时空关联的预测准确度. 现在广泛使用的湍流涡黏亚格子模型的理论基础是能量平衡方程, 它们能够正确预测湍流的空间能量谱, 但不能正确预测湍流的时间能量谱. 本文针对大涡模拟的时间尺度问题, 从时空耦合的观点出发, 综述了作者近年来在时空关联方法方面的研究工作. 理论分析和数值模拟结果表明, 大涡模拟总是高估湍流的时间尺度, 其物理机制是涡黏亚格子模型仅能耗散小尺度能量, 但不能反映亚格子涡对大尺度涡的随机反馈作用. 小尺度湍流会影响颗粒的聚集程度和相对扩散. 建立亚格子湍流的拉格朗日时间尺度模型, 为携带颗粒湍流大涡模拟中基于随机微分方程的颗粒相亚格子模型提供封闭参数, 并用于颗粒湍流扩散的大涡模拟.
其他摘要	This paper reviews the space-time correlation method for large-eddy simulation (LES) of turbulent flows. The method can be used to validate and develop the new subgrid scale models for large-eddy simulation, with the aim to improve the time accuracy in terms of either Eulerian or Lagrangian space-time correlations. The broadly-used eddy-viscosity subgrid scale models are developed based on the turbulent energy budget equations. They can yield correct energy spectra in wavenumber space. However, they cannot correctly produce the energy-spectra in frequency space. It is shown that LES with eddy-viscosity-type SGS models in decaying isotropic turbulence over-predicts time correlations compared with DNS. Several subgrid scale models are compared and lead to the following results: the dynamic Smagorinsky model provides much more accurate predictions than the classic Smagorinsky model and slightly more accurate predictions than the spectral eddy-viscosity model. The multiscale LES using the dynamic Smagorinsky model on the small scale equations is the most accurate one. In turbulent channel flows, the decays of space-time correlations in LES flow field are obvious slower in viscous sublayer and buffer layers than those in DNS, while the results in the log-law region and outer region are similar with DNS. The main physical mechanism is that the subgrid scale models only act as a factitious viscosity to dissipate subgrid scale energy, they cannot modeling the random backscatter of energy from small scales to large ones. For particle-laden turbulent flows, the unresolved small scale in turbulent motions affect the level of clustering of particles and result in an under-prediction of collision rate and relative dispersion of heavy particles at small and moderate Stokes numbers. The model of the subgrid scale Lagrangian timescale accounting for the effects of gravitational settling velocity is established for the closure of particle subgrid scale model based on the stochastic different equation. This model is used to improve the prediction of relative dispersion of heavy particles by turbulent flows.
关键词	湍流大涡模拟时空关联湍流噪声携带颗粒湍流 Turbulent Flows Large-eddy Simulation Space-time Correlation Turbulent Noise Particle-laden Turbulent Flows
DOI	10.1360/SSPMA2015-00446
收录类别	CSCD
语种	中文
CSCD记录号	CSCD:5582789
课题组名称	LNM湍流
论文分区	二类
引用统计
文献类型	期刊论文
条目标识符	http://dspace.imech.ac.cn/handle/311007/58216
专题	非线性力学国家重点实验室
推荐引用方式 GB/T 7714	晋国栋,陈进财,时北极,等. 湍流大涡模拟的时空关联方法[J]. 中国科学: 物理学力学天文学,2015,45,12,:124702.
APA	晋国栋,陈进财,时北极,&何国威.(2015).湍流大涡模拟的时空关联方法.中国科学: 物理学力学天文学,45(12),124702.
MLA	晋国栋,et al."湍流大涡模拟的时空关联方法".中国科学: 物理学力学天文学 45.12(2015):124702.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
湍流大涡模拟的时空关联方法.pdf（3034KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载