首页> 外文学位 >Mixing of scalars in turbulent flows using direct numerical simulations.
【24h】

Mixing of scalars in turbulent flows using direct numerical simulations.

机译:使用直接数值模拟在湍流中混合标量。

获取原文
获取原文并翻译 | 示例

摘要

The research presented in this thesis focuses on scalar mixing in unstratified (neutral) flows and stably stratified flows using Direct Numerical Simulations (DNS). Such flows are ubiquitous in natural flows such as rivers, estuaries, oceans and the atmosphere. First, a detailed study was performed to investigate the effect of varying Schmidt numbers (Sc) on turbulent mixing of a passive scalar in a stationary homogeneous unstratified flow using forced DNS. A total of 6 simulations were performed for 0.1 ≤ Sc < 3. Qualitative and quantitative results of the flow field and the passive scalar fields are presented and discussed. The effect of the Schmidt number on the turbulent mixing was found to be negligible and becomes important (as it should) only when mixing occurs under laminar flow conditions.;Using a model proposed by Venayagamoorthy and Stretch in 2006 for the turbulent diascalar diffusivity as a basis, a practical (and new) model for quantifying the turbulent diascalar diffusivity is proposed as KS=1.1 gamma' LT k1/2, where L T is defined as the Thorpe length scale, k is the turbulent kinetic energy and gamma' is one-half of the mechanical to scalar time scale ratio, which was shown by previous researchers to be approximately 0.7. The novelty of the proposed model lies in the use of LT, which is a widely used length scale in stably stratified flows (almost exclusively used in oceanography), for quantifying turbulent mixing in unstratified flows. LT can be readily obtained in the field using a Conductivity, Temperature and Depth (CTD) profiler or obtained from density fields in a numerical model. The turbulent kinetic energy is mostly contained in the large scales of the flow field and hence can be measured in the field using devices such as an Acoustic Doppler Current Profiler (ADCP) or modeled in numerical simulations. Comparisons using DNS data show remarkably good agreement between the predicted and exact diffusivities.;Finally, the suitability of the proposed model for stably stratified flows was explored for varying degrees of stratification ranging from mildly stable flow conditions to strongly stable conditions. In stably stratified flows, density variations of the fluid dynamically affect the flow field and hence the density acts as what is widely known as an active scalar. Under strongly stable conditions, the DNS results indicate an inverse relationship between the Thorpe scale LT and kinetic energy length scale Lkepsilon, which is different to the direct (almost one to one correspondence) relationship that was found for unstratified flows. Hence, in order to account for this difference, a modified turbulent diascalar diffusivity model was proposed as K d = 13 gamma' LT3 k 1/2. It must be noted that this modified model while dimensionally inconsistent (due to the inverse relationship between the length scales), provides reasonable quantitative estimates of the diffusivity under stably stratified flow conditions.;The models proposed in this study require further (extensive) testing under higher Reynolds number flow conditions. If shown to be valid, they would be widely useful for quantifying turbulent mixing using field measurements of large scale quantities (i.e. LT and k ) as well as a simple and improved turbulence closure scheme.
机译:本文提出的研究集中在使用直接数值模拟(DNS)的非分层(中性)流和稳定分层流中的标量混合。这种流动在河流,河口,海洋和大气等自然流动中无处不在。首先,进行了详细的研究,以研究使用强制DNS在固定的均匀非分层流中,改变施密特数(Sc)对被动标量在湍流混合中的影响。对于0.1≤Sc <3,总共进行了6次仿真。提出并讨论了流场和无源标量场的定性和定量结果。发现Schmidt数对湍流混合的影响可忽略不计,并且仅在层流条件下发生混合时才重要(应如此)。使用Venayagamoorthy和Stretch于2006年提出的模型,将湍流scalscalar扩散率为在此基础上,提出了一种实用的(新的)量化湍流径面扩散系数的模型,其为KS = 1.1 gamma'LT k1 / 2,其中LT定义为索普长度尺度,k是湍流动能,而gamma'是-机械与标量时间比例的一半,以前的研究人员表明约为0.7。所提出模型的新颖性在于LT的使用,它是稳定分层流(几乎专门用于海洋学)中广泛使用的长度标度,用于量化未分层流中的湍流混合。 LT可以使用电导率,温度和深度(CTD)轮廓仪在现场轻松获得,也可以从数值模型中的密度场获得。湍动能主要包含在大范围的流场中,因此可以使用诸如声学多普勒电流剖面仪(ADCP)之类的设备在该场中进行测量或在数值模拟中建模。使用DNS数据进行的比较显示出预测的扩散率和精确的扩散率之间的一致性非常好。最后,探讨了该模型对稳定分层流的适用性,适用于从中等程度稳定的流动条件到强烈稳定条件的各种分层程度。在稳定的分层流中,流体的密度变化会动态影响流场,因此密度起着众所周知的主动标量的作用。在高度稳定的条件下,DNS结果表明索普尺度LT与动能长度尺度Lkepsilon之间存在反比关系,这与非分层流的直接(几乎一对一)关系不同。因此,为了解决这一差异,提出了一种改进的湍流径面扩散率模型,其模型为K d = 13 gamma'LT3 k 1/2。必须注意的是,这种修改后的模型虽然尺寸不一致(由于长度刻度之间存在反比关系),但可以在稳定的分层流动条件下提供合理的定量扩散系数估计值;本研究中提出的模型需要在以下条件下进行进一步的(广泛的)测试雷诺数更高的流动条件。如果被证明是有效的,它们将对使用大规模量(即LT和k)的现场测量以及简单而改进的湍流闭合方案来量化湍流混合非常有用。

著录项

  • 作者单位

    Colorado State University.;

  • 授予单位 Colorado State University.;
  • 学科 Civil engineering.
  • 学位 M.S.
  • 年度 2015
  • 页码 77 p.
  • 总页数 77
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

  • 入库时间 2022-08-17 11:52:47

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号