CFD中文网

hoversoar

@Zhy2022 我之前稳定了之后从1调到过5和10，结果没什么区别。仅供参考。

hoversoar

重新画了mesh，模型改成了sst。Re 5w
结果更奇怪了，直接算不下去。下图无量纲时间步

现在mesh的y+平均1.5，max 3，min0.11。应该是没什么问题，质量也很高。
上下cyclicAMI边界，圆形流域，spanwise piD。
没想到一个RANS花了这么多天还没算好。。
有大佬能帮我看看么

k

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 0.0021;

boundaryField
{
    front
    {
        type            cyclicAMI;
        value           uniform 0.0021;
    }
    back
    {
        type            cyclicAMI;
        value           uniform 0.0021;
    }
    inlet
    {
        type            fixedValue;
        value           uniform 0.0021;
    }
    outlet
    {
        type            inletOutlet;
        inletValue      internalField;
        value           $internalField;
    }
    cylinderwalls
    {
        type            kqRWallFunction;
        value           uniform 0.0021;
    }
}

omega

dimensions      [ 0 0 -1 0 0 0 0 ];

internalField   uniform 10.9;

boundaryField
{
    cylinderwalls
    {
        type            omegaWallFunction;
        value           $internalField;
    }
    front
    {
        type            cyclicAMI;
    }
    back
    {
        type            cyclicAMI;
    }
    inlet
    {
        type            fixedValue;
        value           uniform 10.9;
    }
    outlet
    {
        type            inletOutlet;
        inletValue      internalField;
        value           $internalField;
    }
}

nut

dimensions      [0 2 -1 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    inlet
    {
        type            calculated;
        value           uniform 0;
    }
    outlet
    {
        type            calculated;
        value           uniform 0;
    }
    cylinderwalls
    {
        type            nutkWallFunction;
        value           uniform 0;
    }
    front
    {
        type            cyclicAMI;
    }
    back
    {
        type            cyclicAMI;
    }
}

p

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    front
    {
        type            cyclicAMI;
        value           uniform 0;
    }
    back
    {
        type            cyclicAMI;
        value           uniform 0;
    }
    inlet
    {
        type            zeroGradient;
    }
    outlet
    {
        type            fixedValue;
        value           uniform 0;
    }
    cylinderwalls
    {
        type            zeroGradient;
    }
}

U

dimensions      [0 1 -1 0 0 0 0];

internalField   uniform (7.5 0 0);

boundaryField
{
    front
    {
        type            cyclicAMI;
        value           uniform (7.5 0 0);
    }
    back
    {
        type            cyclicAMI;
        value           uniform (7.5 0 0);
    }
    inlet
    {
        type            fixedValue;
        value           uniform (7.5 0 0);
    }
    outlet
    {
        type            zeroGradient;
    }
    cylinderwalls
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }
}

fvscheme

ddtSchemes
{
    default         backward;
}

gradSchemes
{
    default         Gauss linear;
    grad(p)         Gauss linear;
    grad(U)         Gauss linear;
}

divSchemes
{
    default         none;
    div(phi,U)      Gauss limitedLinearV 0.5;
    div(phi,k)      Gauss limitedLinear 1;
    div(phi,omega) Gauss limitedLinear 1;
    div(phi,R)      Gauss limitedLinear 1;
    div(R)          Gauss linear;
    div(phi,B)      Gauss limitedLinear 1;
    div(B)          Gauss linear;
    div(phi,nuTilda) Gauss limitedLinear 1;
    div((nuEff*dev(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
    default         Gauss linear corrected;
}

interpolationSchemes
{
    default         linear;
    interpolate(U)  linear;
}

snGradSchemes
{
    default         corrected;
}

fluxRequired
{
    default         no;
    p;
    Phi;
}

fvsolution

solvers
{
    p
    {
        solver          GAMG;
        tolerance       1e-04;
        relTol          0;
        smoother        GaussSeidel;
        cacheAgglomeration true;
        nCellsInCoarsestLevel 200;
        agglomerator    faceAreaPair;
        mergeLevels     1;
    }

    pFinal
    {
        solver          GAMG;
        tolerance       1e-06;
        relTol          0;
        smoother        GaussSeidel;
        cacheAgglomeration true;
        nCellsInCoarsestLevel 200;
        agglomerator    faceAreaPair;
        mergeLevels     1;
    }

    "(U|k|nut|omega)"
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-05;
        relTol          0;
    }

    "(U|k|nut|omega)Final"
    {
        $U;
        tolerance       1e-06;
        relTol          0;
    }
}

PIMPLE
{
    nOuterCorrectors 3;
    nCorrectors     2;
    nNonOrthogonalCorrectors 1;
    pRefCell        0;
    pRefValue       0;
}

hoversoar

更新一下
算了300个无量纲时间步，Cd-0.6，Clrms-0.31.

在15w雷诺数的时候应该还没有到临界，Cd应该在1-1.2的量级。
不知道是不是y+在背风面过小的问题（2.5左右）。
我加大了第一层网格，0.03D重新算。

hoversoar

之前一直用LES, 现在尝试RANS，有一些问题希望和大家探讨一下。
现在再算的算例基本概况：三维圆柱绕流，雷诺数15w，现在采用RNG k-epsilon模型，结构化网格质量较高。
虽然刚算了一会但是对于RANS的y+取值有一些困惑。

网上基本上是说k-e模型的y+在30-200之间最好，要大于11。这个数值是针对的最大值，最小值，还是平均值呢？

查看了最近的时间步的y+，数值如下。可以看到最小值和最大值差的很远，最小值也是落在了粘性底层的范围内。（个人认为可能小范围的偏差不要紧，只要平均值满足即可，于是我查看了y+的分布）

Patch 2 named cylinderwalls y+ : min: 2.63471 max: 66.0352 average: 37.2911

y+分布如下。来流为x+方向。可以看到分离之后大片区域是在10以下的。

一开始打算算的是5w的例子，结果发现用该湍流模型min y+很难增大到所谓的30。于是同时调整了雷诺数和网格厚度，现在的网格首层已经达到0.01D和15w雷诺，觉得已经很粗糙了。可是y+结果还是偏差比较大。

采用k-e算例是会使得y+难以增大到11以上吗，还是所谓的高雷诺数指的是相当量级的值（譬如50w）？

附上现在的算例文档，有感兴趣的同学可以一起探讨下。suanli.zip

hoversoar

我后来的做法是将网格做到y+<1，然后不采用壁面函数，k方程壁面边界条件为kqRwallFuction, omega边界条件为omegaWallFuction, nut边界条件为nutkWallFunction
另外，我觉得将网格做到y+>30，然后指定壁面函数也是可以的

同样有疑问，指定了k w 和nut 的边界条件不就是采用了壁面函数吗？
还是说壁面函数有什么另外的指定方法？
（还是我认知错误）

hoversoar

@random_ran OK 感谢! 很好的分享！

hoversoar

鼓掌，斯国一。非常感谢你的分享。
结果看来是对流格式得影响最大。
我之前对Re1w的圆柱绕流也是进行了很多尝试，主要是对流项的。为什么你这边要用linearUpwind作为基准呢？
我的网格质量由于不能做到很好，所以尝试的是limitedLinear，以及调整后面的参数。分别用了0.2，0.5和1。但是后面的系数1的话重视安定性。0的话重视精度，如果在质量很好的网格下，0的结果应该是接近语linear的。
所以个人觉得并不是要避免使用limitedLinear模型，在网格质量不是太好的情况下，大家可以尝试这个系数的调整来达到精度和安定的平衡。（一点拙见）

hoversoar

@红豆沙你是不能计算还是啥问题啊。你导进去OF里头试看看。

hoversoar

@kisssinger 看运行了几个核就知道有没有在运行了。或者terminal 打top查看。

hoversoar

报告一下，之前也算过dynamicKEqn并没有问题，检查了老半天。拿到之前另一个电脑算就可以运行。。
在这个报错的of5的电脑上安装了of7，现在可以顺利地运行了。

hoversoar

@hoversoar 应该是滤波的问题？还不确定为什么会产生...

hoversoar

请问这个滤波为什么会报错有什么结果吗? 我现在也碰到报错了，但是其他mesh又可以用，在怀疑是不是mesh的问题？

hoversoar

尝试把湍流模型从kEqn换成了dynamicKEqn，结果报了错。没看懂，有大佬能帮我解释一下吗..

[10] #0  Foam::error::printStack(Foam::Ostream&)[12] #0  Foam::error::printStack(Foam::Ostream&)[13] #0  [7] #0  Foam::error::printStack(Foam::Ostream&)Foam::error::printStack(Foam::Ostream&)[3] #0  [11] #0  Foam::error::printStack(Foam::Ostream&)Foam::error::printStack(Foam::Ostream&) at ??:?
[12] #1  Foam::sigFpe::sigHandler(int) at ??:?
 at ??:?
[11] #1  Foam::sigFpe::sigHandler(int)[10] #1  Foam::sigFpe::sigHandler(int) at ??:?
[3] #1  Foam::sigFpe::sigHandler(int) at ??:?
[13] #1  Foam::sigFpe::sigHandler(int) at ??:?
[7] #1  Foam::sigFpe::sigHandler(int) at ??:?
[12] #2  ? at ??:?
[3] #2  ? at ??:?
[11] #2  ? at ??:?
[10] #2  ? at ??:?
[7] #2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
[12] #3  Foam::sqrt(Foam::Field<double>&, Foam::UList<double> const&) at ??:?
[13] #2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
 in "/lib/x86_64-linux-gnu/libc.so.6"
[3] #3  Foam::sqrt(Foam::Field<double>&, Foam::UList<double> const&) in "/lib/x86_64-linux-gnu/libc.so.6"
[10] #3  Foam::sqrt(Foam::Field<double>&, Foam::UList<double> const&)[7] #3  Foam::sqrt(Foam::Field<double>&, Foam::UList<double> const&) in "/lib/x86_64-linux-gnu/libc.so.6"
[11] #3  Foam::sqrt(Foam::Field<double>&, Foam::UList<double> const&) in "/lib/x86_64-linux-gnu/libc.so.6"
[13] #3  Foam::sqrt(Foam::Field<double>&, Foam::UList<double> const&) at ??:?
[12] #4  Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam::sqrt<Foam::fvPatchField, Foam::volMesh>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[3] #4  Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam::sqrt<Foam::fvPatchField, Foam::volMesh>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[10] #4  Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam::sqrt<Foam::fvPatchField, Foam::volMesh>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[7] #4  Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam::sqrt<Foam::fvPatchField, Foam::volMesh>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[11] #4  Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam::sqrt<Foam::fvPatchField, Foam::volMesh>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[12] #5  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::Ck(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
[13] #4  Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam::sqrt<Foam::fvPatchField, Foam::volMesh>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[3] #5  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::Ck(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
[10] #5  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::Ck(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
[11] #5  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::Ck(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
[7] #5  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::Ck(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
[12] #6  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[10] #6  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[13] #5  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::Ck(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
[12] #7  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut() at ??:?
[3] #6  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[11] #6  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[7] #6  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[13] #6  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut(Foam::GeometricField<Foam::SymmTensor<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[12] #8   at ??:?
[10] #7  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut() at ??:?
[11] #7  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut() at ??:?
[3] #7  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut() at ??:?
[7] #7  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut()? at ??:?
[13] #7  Foam::LESModels::dynamicKEqn<Foam::IncompressibleTurbulenceModel<Foam::transportModel> >::correctNut() at ??:?
[11] #8   at ??:?
[12] #9  __libc_start_main at ??:?
[10] #8   at ??:?
[13] #8   at ??:?
[7] #8   at ??:?
[3] #8  ?? in "/lib/x86_64-linux-gnu/libc.so.6"
[12] #10  ?? at ??:?
[11] #9  __libc_start_main at ??:?
[13] #9  __libc_start_main?? in "/lib/x86_64-linux-gnu/libc.so.6"
[13] #10   in "/lib/x86_64-linux-gnu/libc.so.6"
[11] #10   at ??:?
[10] #9  __libc_start_main at ??:?
[3] #9  __libc_start_main at ??:?
[7] #9  __libc_start_main?? in "/lib/x86_64-linux-gnu/libc.so.6"
[10] #10   at ??:?
[DL-BoxII:41641] *** Process received signal ***
[DL-BoxII:41641] Signal: Floating point exception (8)
[DL-BoxII:41641] Signal code:  (-6)
[DL-BoxII:41641] Failing at address: 0x3ea0000a2a9
[DL-BoxII:41641] [ 0] /lib/x86_64-linux-gnu/libc.so.6(+0x354b0)[0x2ad9e28f14b0]
[DL-BoxII:41641] [ 1] /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x38)[0x2ad9e28f1428]
[DL-BoxII:41641] [ 2]  in "/lib/x86_64-linux-gnu/libc.so.6"
[3] #10  /lib/x86_64-linux-gnu/libc.so.6(+0x354b0)[0x2ad9e28f14b0]
[DL-BoxII:41641] [ 3] /opt/openfoam5/platforms/linux64GccDPInt32Opt/lib/libOpenFOAM.so(_ZN4Foam4sqrtERNS_5FieldIdEERKNS_5UListIdEE+0x28)[0x2ad9e1941728]
[DL-BoxII:41641] [ 4] /opt/openfoam5/platforms/linux64GccDPInt32Opt/lib/libincompressibleTurbulenceModels.so(_ZN4Foam4sqrtINS_12fvPatchFieldENS_7volMeshEEENS_3tmpINS_14GeometricFieldIdT_T0_EEEERKS7_+0x190)[0x2ad9ddd760f0]
[DL-BoxII:41641] [ 5] /opt/openfoam5/platforms/linux64GccDPInt32Opt/lib/libincompressibleTurbulenceModels.so(_ZNK4Foam9LESModels11dynamicKEqnINS_29IncompressibleTurbulenceModelINS_14transportModelEEEE2CkERKNS_14GeometricFieldINS_10SymmTensorIdEENS_12fvPatchFieldENS_7volMeshEEERKNS6_IdS9_SA_EE+0x2a5)[0x2ad9ddd87f15]
[DL-BoxII:41641] [ 6] /opt/openfoam5/platforms/linux64GccDPInt32Opt/lib/libincompressibleTurbulenceModels.so(_ZN4Foam9LESModels11dynamicKEqnINS_29IncompressibleTurbulenceModelINS_14transportModelEEEE10correctNutERKNS_14GeometricFieldINS_10SymmTensorIdEENS_12fvPatchFieldENS_7volMeshEEERKNS6_IdS9_SA_EE+0x79)[0x2ad9ddd88859]
[DL-BoxII:41641] [ 7] /opt/openfoam5/platforms/linux64GccDPInt32Opt/lib/libincompressibleTurbulenceModels.so(_ZN4Foam9LESModels11dynamicKEqnINS_29IncompressibleTurbulenceModelINS_14transportModelEEEE10correctNutEv+0x1f4)[0x2ad9ddd89874]
[DL-BoxII:41641] [ 8] pimpleFoam[0x4254ad]
[DL-BoxII:41641] [ 9] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0)[0x2ad9e28dc830]
[DL-BoxII:41641] [10] pimpleFoam[0x428729]
[DL-BoxII:41641] *** End of error message ***
 at ??:?

hoversoar

@东岳完全失败中...

hoversoar

@cccrrryyy 感谢解答。

我看了pimpleFoam的这个算例，不太清楚具体是哪里用到了B，去掉的话可以算么？按道理说它用的是WALE模型，是个代数模型，不需要用方程去求解B才对。pisoFoam用的是SpalartAllmaras模型，不涉及到求解B，但要求解nuTilda。不太清楚为什么需要div(phi,k)，可能这个去掉也可以算。

pimpleFoam没有B项也是可以算的，我这期间就是没发现还有个这个就一直在算..所以才会疑惑。

hoversoar

大家好，又有了新的困惑。
我翻看教程的时候，发现pisofoam和pimplefoam里头LES案例的fvscheme 对流项设置有一些不同，觉得有点困惑。pimplefoam里头有一个B项，pisofoam里头则没有。
pimplefoam/LES/channel395

divSchemes
{
    default         none;
    div(phi,U)      Gauss linear;
    div(phi,k)      Gauss limitedLinear 1;
    div(phi,B)      Gauss limitedLinear 1;
    div(B)          Gauss linear;
    div(phi,nuTilda) Gauss limitedLinear 1;
    div((nuEff*dev2(T(grad(U))))) Gauss linear;
}

pisofoam/LES/motobike:

divSchemes
{
    default         none;
    div(phi,U)      Gauss LUST unlimitedGrad(U);
    div(phi,k)      Gauss limitedLinear 1;
    div(phi,nuTilda) Gauss limitedLinear 1;
    div((nuEff*dev2(T(grad(U))))) Gauss linear;
}

查了下LES对应的B，RAS用的R，是雷诺应力张量的求解。那既然default都是none，按照user guide是要对每一项进行定义的，为什么pisofoam里头缺失了这一项呢？
ps我看到我之前的case也都没写这一项，估计是直接复制了摩托车的代码。不知道这项是不是必须的？

hoversoar

@cccrrryyy

从这个角度讲，感觉可能你需要的是花大量精力在网格上，而不是求解器的那些参数调整。

之前几个月自己就一顿乱算，导致连光滑圆柱的结果都很奇怪。所以用那个简单的模型在探讨求解器，设置参数之类的。
对得上了之后，再过来讨论这边这个圆柱的网格。

你之前提到过跟现有文献对比，是哪个文章呀，挺感兴趣的。

非光滑圆柱的话，参考的Experimental study on flow past a circular cylinder with rough surface这篇文献的。

hoversoar

@羽之下还可以打开链接啊。
再贴一次。
壁面函数

hoversoar

@东岳谢谢夸奖！
算是基于工程的研究吧.. 桥索上用的

hoversoar

@cccrrryyy

感觉ANSYS Meshing的思路是尽可能缩小五边形所占的比例，从而大量使用六边形？

这个倒是没有研究, 就让他自动转化了。个人觉得可能形状没有太拘泥，如果不是edge的处不能合并三角形的话，尽量会合并成多边形并且满足所设定的size（也就是不会合并到过分大）。

CFD中文网

hoversoar

帖子