首页> 外文会议>Annual conference of the Canadian Society for Civil Engineering >A New Mesh Refinement Indicator for Depth Averaged Modeling of Shallow Flow with Variable Bed Topography
【24h】

A New Mesh Refinement Indicator for Depth Averaged Modeling of Shallow Flow with Variable Bed Topography

机译:一种新的网格细化指标,用于可变床面地形的浅流深度平均建模

获取原文

摘要

Mountainous streams are characterized by highly variable bed topography, steep gradients, and small depths. As a result, modeling of the flow in these channels can be challenging. The accuracy and stability of the model depend on the discretization, but there are little existing criteria or guidelines for specifying the discretization. When the depth becomes small compared to the discretization scale, the source terms generally dominate in the Saint-Venant equations. However local variations can lead to sub-critical/supercritical transitions necessitating a conservative upwind shock-capturing numerical scheme. A previous Fourier analysis of such a scheme applied to the linearized non-dimensional 1D Saint-Venant equations showed that the oscillations in the steady state solution depend on the amplitude of the bed profile perturbation and the ratio of the discretization scale to a characteristic length scale. The characteristic length scale is defined as the ratio of the uniform flow depth to the average bed slope. In this paper, the definition of the slope and the depth in the parameters and the utility of the parameters are investigated in the case of two-dimensional (2D) finite element flow model. The last non-dimensional parameter is used as a mesh refinement indicator with an idealized test case, flow past a submerged groin. The parameters are valid for any numerical scheme and are relatively easy to calculate and implement. The analysis will lead us to a stable and accurate hydrodynamic solution which will help with better solutions to other models, e.g. morphology models, transport models, habitat models, ice models etc.
机译:山区河流的特征是床层地形高度可变,陡峭的坡度和较小的深度。结果,对这些通道中的流进行建模可能具有挑战性。模型的准确性和稳定性取决于离散化,但是很少有用于指定离散化的现有标准或准则。与离散化尺度相比,当深度变小时,源项通常在Saint-Venant方程中占主导地位。然而,局部变化会导致亚临界/超临界转变,因此需要采用保守的迎风捕捉数值方案。先前将这种方案应用于线性化的无量纲一维Saint-Venant方程的傅立叶分析表明,稳态解中的振动取决于床轮廓扰动的幅度以及离散化比例与特征长度比例的比值。特征长度标度定义为均匀流动深度与平均床坡度之比。本文研究了二维(2D)有限元流模型中参数的斜率和深度的定义以及参数的效用。最后一个无量纲参数用作理想化测试用例的网格细化指标,流经浸没式腹股沟。这些参数对于任何数值方案均有效,并且相对容易计算和实现。该分析将使我们找到一个稳定而准确的流体动力解决方案,这将有助于更好地解决其他模型的问题,例如:形态模型,运输模型,栖息地模型,冰模型等

著录项

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