LES of a classical hydraulic jump

This dataset contains results from Large-Eddy Simulations of a classical hydraulic jump at Fr=2. It accompanies the following article:<br>T. Mukha, S. K. Almeland, and R. E. Bensow. LES of a classical hydraulic jump: Influence of modelling parameters on the predictive accuracy. 2020.<br><br>Kindly, cite the article if you make use of the data.<br>The data for each case in the article is found in the associated zip archive. For example, for the simulations using MULES, dx2 and 10% upwinding, the name is mules_dx2_lu10.zip.Inside each such zip file there is the following:- A .vtm file containing full averaged 2d fields. Can be opened with paraview or turbulucid.<br>- The r_z_* files containing two-point correlations in the spanwise direction.- The spectra directory containing the computed time-spectra.- The probes_* files, containing the probes data which was used to compute the spectra and the acfs.- The probes_resampled_* files, containing the same probe data but resampled to evenly distributed time-values, with 0.001 s between the samples.<br>The leaked_air.hdf5 file contains the sampled signals for the volume of leaked air, for all cases.<br>The leaked_air_resampled.hdf5 contains the same data but resampled to evenly distributed time-values, with 0.001 s between the samples.<br>The cases archive contains OpenFOAM setup files for simulations on all four meshes. In the paper, v1806 is used, but probably later versions should work as well. The function objects inside rely on swak4foam.<br>To run the cases, blockMesh should first be used to construct the mesh followed by "topoSet dict system/swak.topoSetDict" to define the cellSet used for sampling the volume of leaked air. Case dx1 is special due to the additional mesh refinement step.Here, first run blockmesh, then "mv 2 2.ogr", "topoSet dict system/topoSetDict_ref", "refineHexMesh -overwrite refine" "mv 2.org 2"Be prepared that the mesh refinment will take a long time, possibly hours.<br>To reproduce a particular case, adjust the settings with respect to the amount of upwinding and then run interFoam for MULES and interIsoFoam for the isoAdvector VoF. <br>The two interface_animation_data archives contains a series of VTK files containing the air-water interface surface, which can be used to look at the interface dynamics. The data is extracted from the benchmark simulation.<br><br>Finally, the videos archive contains rendered animations of the vtk files.<br><br>本数据集包含弗劳德数Fr=2的经典水跃的大涡模拟（Large-Eddy Simulations, LES）结果，配套发表于以下论文：<br>T. Mukha, S. K. Almeland 与 R. E. Bensow. 经典水跃的大涡模拟：建模参数对预测精度的影响. 2020.<br><br>若您使用本数据集，请务必引用上述论文。<br>论文中各工况的数据均存储于配套的压缩归档文件中。例如，采用MULES格式、空间步长dx2及10%迎风格式的模拟对应归档文件名为`mules_dx2_lu10.zip`。<br>每个此类压缩包内包含以下内容：<br>- 一个.vtm格式文件，存储完整的二维平均场数据，可通过ParaView或Turbulucid打开；<br>- `r_z_*`格式文件，存储展向的两点相关函数数据；<br>- `spectra`目录，存储计算得到的时间频谱数据；<br>- `probes_*`格式文件，存储用于计算频谱与自相关函数（ACFs）的探针测量数据；<br>- `probes_resampled_*`格式文件，存储经重采样后的探针数据，采样时间间隔为0.001秒，时间序列均匀分布。<br><br>`leaked_air.hdf5`文件存储了所有工况下泄漏空气体积的采样信号。<br>`leaked_air_resampled.hdf5`则存储了经重采样后的上述数据，采样时间间隔同样为0.001秒。<br><br>`cases`压缩归档包含了四种网格的模拟OpenFOAM配置文件。论文中使用的是OpenFOAM v1806版本，但后续版本大概率也可兼容。其中的函数对象依赖于swak4foam库。<br><br>运行工况的步骤如下：首先需通过`blockMesh`命令构建网格，随后执行`topoSet dict system/swak.topoSetDict`命令，定义用于采样泄漏空气体积的单元集。<br>其中工况dx1较为特殊，包含额外的网格细化步骤：需先运行`blockMesh`，再依次执行`mv 2 2.ogr`、`topoSet dict system/topoSetDict_ref`、`refineHexMesh -overwrite refine`以及`mv 2.org 2`。请注意，网格细化过程耗时较长，可能需要数小时。<br><br>若需复现特定工况，请根据迎风格式占比调整相关设置：对于采用MULES的模拟，请运行`interFoam`命令；对于采用isoAdvector流体体积（Volume of Fluid, VoF）方法的模拟，请运行`interIsoFoam`命令。<br><br>两个`interface_animation_data`压缩归档包含一系列VTK格式文件，存储气水交界面的表面数据，可用于观察交界面的动态变化。该数据提取自基准模拟。<br><br>最后，`videos`压缩归档包含上述VTK文件的渲染动画。