OpenFOAM中cyclicAMI周期边界在有边界层时出现问题

Foamer24

各位专家好，小弟最近用OpenFOAM计算三维圆柱绕流，提供了两个算例，一个可算的例子适合新手入门；一个出错的例子，出错原因仅仅因为将最小网格高度从第一个例子的0.001m变为0.00001m。这种对靠近壁面的边界层网格加密操作，在工程中经常需要，难以避免，烦请各位大神指教。

第一个例子的第一层网格高度（即最小网格高度）为0.001m，可以正常计算，该算例地址为 https://www.jianguoyun.com/p/DQL0lW8QgOLABxjSjroB

详细参数如下：

边界条件：圆柱wall，圆柱两端cyclicAMI周期边界，两周期边界间距为1，其余为远场边界patch；
求解器：可压缩流求解器HISA（使用git clone https://gitlab.com/hisa/hisa.git 下载，用./Allwmake安装）；
网格绘制：Pointwise（能算的网格，最小网格高度0.001m；不能算的网格，最小高度0.00001m）。

具体操作如下：

//注意边界条件已设好，若未设好需额外用createPatch运行system/createPatchDict
//网格尺寸缩放，这里已经缩放完了，所以都是1
transformPoints -scalle '(1 1 1)'
//网格重新排序
renumberMesh -overwrite
//添加周期边界
setSet
faceSet  AMI  new  patchToFace  wall1
faceSet  AMI  add  patchToFace  wall2
quit
setsToZones –noFlipMap
//开始单核串行计算
hisa
//开始多核并行计算
mpirun -np 4 hisa -parallel

第二个不能正常计算的算例，仅在同一套网格中加密第一层网格高度到0.00001m，算例设置同上，地址为 https://www.jianguoyun.com/p/DYRDSoMQgOLABxjxjroB 。在不使用renumberMesh -overwrite操作时，可以单核串行计算，但是并行时会出现下面的错误：

/*---------------------------------------------------------------------------*\
  =========                 |
  \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
   \\    /   O peration     | Website:  https://openfoam.org
    \\  /    A nd           | Version:  6
     \\/     M anipulation  |
\*---------------------------------------------------------------------------*/
Build  : 6-76a8284120c3
Exec   : hisa -parallel
Date   : May 02 2019
Time   : 08:38:04
Host   : "localhost.localdomain"
PID    : 37453
I/O    : uncollated
Case   : /home/xters/Desktop/cylinder_bl
nProcs : 4
Slaves : 
3
(
"localhost.localdomain.37454"
"localhost.localdomain.37455"
"localhost.localdomain.37456"
)

Pstream initialized with:
    floatTransfer      : 0
    nProcsSimpleSum    : 0
    commsType          : nonBlocking
    polling iterations : 0
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster (fileModificationSkew 10)
allowSystemOperations : Allowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Overriding DebugSwitches according to controlDict
    cyclicAMI       ;

Create mesh for time = 0, region region0
Creating static mesh

[1] cyclicAMIPolyPatch::calcTransforms : patch:wall1 Specified translation : (0 0 -1)
[1] patch: wall1
[1]     forwardT = 0()
[1]     reverseT = 0()
[1]     separation = 1((0 0 -1))
[1]     collocated = 1(0)
[1] 
[1] cyclicAMIPolyPatch::calcTransforms : patch:wall2 Specified translation : (0 0 1)
[1] patch: wall2
[1]     forwardT = 0()
[1]     reverseT = 0()
[1]     separation = 1((0 0 1))
[1]     collocated = 1(0)
[1] 
[2] cyclicAMIPolyPatch::calcTransforms : patch:wall1 Specified translation : (0 0 -1)
[2] patch: wall1
[2]     forwardT = 0()
[2]     reverseT = 0()
[2]     separation = 1((0 0 -1))
[2]     collocated = 1(0)
[2] 
[2] cyclicAMIPolyPatch::calcTransforms : patch:wall2 Specified translation : (0 0 1)
[2] patch: wall2
[2]     forwardT = 0()
[2]     reverseT = 0()
[2]     separation = 1((0 0 1))
[2]     collocated = 1(0)
[2] 
[3] cyclicAMIPolyPatch::calcTransforms : patch:wall1 Specified translation : (0 0 -1)
[3] patch: wall1
[3]     forwardT = 0()
[3]     reverseT = 0()
[3]     separation = 1((0 0 -1))
[3]     collocated = 1(0)
[3] 
[3] cyclicAMIPolyPatch::calcTransforms : patch:wall2 Specified translation : (0 0 1)
[3] patch: wall2
[3]     forwardT = 0()
[3]     reverseT = 0()
[3]     separation = 1((0 0 1))
[3]     collocated = 1(0)
[3] 
[0] cyclicAMIPolyPatch::calcTransforms : patch:wall1 Specified translation : (0 0 -1)
[0] patch: wall1
[0]     forwardT = 0()
[0]     reverseT = 0()
[0]     separation = 1((0 0 -1))
[0]     collocated = 1(0)
[0] 
[0] cyclicAMIPolyPatch::calcTransforms : patch:wall2 Specified translation : (0 0 1)
[0] patch: wall2
[0]     forwardT = 0()
[0]     reverseT = 0()
[0]     separation = 1((0 0 1))
[0]     collocated = 1(0)
[0] 
Steady-state analysis detected

pseudoTime: tolerance = 1e-06 (1e-06 1e-06 1e-06) 1e-06 , relTol = 0 (0 0 0) 0 

Local timestepping selected

Reading thermophysical properties

Selecting thermodynamics package 
{
    type            hePsiThermo;
    mixture         pureMixture;
    transport       sutherland;
    thermo          hConst;
    equationOfState perfectGas;
    specie          specie;
    energy          sensibleInternalEnergy;
}

AMI: Creating addressing and weights between 39880 source faces and 39880 target faces
[3] cyclicAMIPolyPatch : wall1 constructed AMI with 
[3]     srcAddress:9980
[3]     tgAddress :9960
[3] 
[1] cyclicAMIPolyPatch : wall1 constructed AMI with 
[1]     srcAddress:9980
[1]     tgAddress :9960
[1] 
[2] cyclicAMIPolyPatch : wall1 constructed AMI with 
[2]     srcAddress:9960
[2]     tgAddress :9980
[2] 
AMI: Patch source sum(weights) min/max/average = 1, 1, 1
AMI: Patch target sum(weights) min/max/average = 1, 1, 1
[0] cyclicAMIPolyPatch : wall1 constructed AMI with 
[0]     srcAddress:9960
[0]     tgAddress :9980
[0] 

Viscous analysis detected

Reading field U

Creating turbulence model

Selecting turbulence model type RAS
Selecting RAS turbulence model SpalartAllmaras
Selecting patchDistMethod meshWave
RAS
{
    RASModel        SpalartAllmaras;
    turbulence      on;
    printCoeffs     on;
    sigmaNut        0.66666;
    kappa           0.41;
    Cb1             0.1355;
    Cb2             0.622;
    Cw2             0.3;
    Cw3             2;
    Cv1             7.1;
    Cs              0.3;
}


Creating flux scheme

Selecting flux scheme AUSMPlusUp

Initial pseudo Courant No: Min: 0.001 Mean: 0.001 Max: 0.001


Starting pseudotime iteration loop

Time = 1

pseudoTime: iteration 1
Pseudo Courant No: Min: 0.001 Mean: 0.001 Max: 0.001
Pseudo deltaT: Min: 1.31771e-11 Mean: 4.2537e-08 Max: 4.91626e-08
Solving for (rho rhoU rhoE)
[localhost:37456] *** An error occurred in MPI_Waitall
[localhost:37456] *** reported by process [1120337921,3]
[localhost:37456] *** on communicator MPI COMMUNICATOR 3 SPLIT FROM 0
[localhost:37456] *** MPI_ERR_TRUNCATE: message truncated
[localhost:37456] *** MPI_ERRORS_ARE_FATAL (processes in this communicator will now abort,
[localhost:37456] ***    and potentially your MPI job)
[localhost.localdomain:37448] 3 more processes have sent help message help-mpi-errors.txt / mpi_errors_are_fatal
[localhost.localdomain:37448] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages

使用renumberMesh -overwrite操作后，单核也会出错了，错误信息为：

/*---------------------------------------------------------------------------*\
  =========                 |
  \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
   \\    /   O peration     | Website:  https://openfoam.org
    \\  /    A nd           | Version:  6
     \\/     M anipulation  |
\*---------------------------------------------------------------------------*/
Build  : 6-76a8284120c3
Exec   : hisa
Date   : May 02 2019
Time   : 08:40:54
Host   : "localhost.localdomain"
PID    : 37564
I/O    : uncollated
Case   : /home/xters/Desktop/cylinder_bl
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster (fileModificationSkew 10)
allowSystemOperations : Allowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Overriding DebugSwitches according to controlDict
    cyclicAMI       ;

Create mesh for time = 0, region region0
Creating static mesh

cyclicAMIPolyPatch::calcTransforms : patch:wall1 Specified translation : (0 0 -1)
patch: wall1
    forwardT = 0()
    reverseT = 0()
    separation = 1((0 0 -1))
    collocated = 1(0)

cyclicAMIPolyPatch::calcTransforms : patch:wall2 Specified translation : (0 0 1)
patch: wall2
    forwardT = 0()
    reverseT = 0()
    separation = 1((0 0 1))
    collocated = 1(0)

Steady-state analysis detected

pseudoTime: tolerance = 1e-06 (1e-06 1e-06 1e-06) 1e-06 , relTol = 0 (0 0 0) 0 

Local timestepping selected

Reading thermophysical properties

Selecting thermodynamics package 
{
    type            hePsiThermo;
    mixture         pureMixture;
    transport       sutherland;
    thermo          hConst;
    equationOfState perfectGas;
    specie          specie;
    energy          sensibleInternalEnergy;
}

#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigFpe::sigHandler(int) at ??:?
#2  ? in "/lib64/libc.so.6"
#3  Foam::surfaceInterpolation::makeWeights() const at ??:?
#4  Foam::surfaceInterpolation::weights() const at ??:?
#5  Foam::surfaceInterpolation::makeDeltaCoeffs() const at ??:?
#6  Foam::surfaceInterpolation::deltaCoeffs() const at ??:?
#7  Foam::fvPatch::deltaCoeffs() const at ??:?
#8  Foam::fvPatchField<double>::snGrad() const at ??:?
#9  Foam::heThermo<Foam::psiThermo, Foam::pureMixture<Foam::sutherlandTransport<Foam::species::thermo<Foam::hConstThermo<Foam::perfectGas<Foam::specie> >, Foam::sensibleInternalEnergy> > > >::heBoundaryCorrection(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh>&) at ??:?
#10  Foam::heThermo<Foam::psiThermo, Foam::pureMixture<Foam::sutherlandTransport<Foam::species::thermo<Foam::hConstThermo<Foam::perfectGas<Foam::specie> >, Foam::sensibleInternalEnergy> > > >::heThermo(Foam::fvMesh const&, Foam::word const&) at ??:?
#11  Foam::psiThermo::addfvMeshConstructorToTable<Foam::hePsiThermo<Foam::psiThermo, Foam::pureMixture<Foam::sutherlandTransport<Foam::species::thermo<Foam::hConstThermo<Foam::perfectGas<Foam::specie> >, Foam::sensibleInternalEnergy> > > > >::New(Foam::fvMesh const&, Foam::word const&) at ??:?
#12  Foam::autoPtr<Foam::psiThermo> Foam::basicThermo::New<Foam::psiThermo>(Foam::fvMesh const&, Foam::word const&) at ??:?
#13  Foam::psiThermo::New(Foam::fvMesh const&, Foam::word const&) at ??:?
#14  Foam::hisaModule::initialise() at ~/Desktop/hisa/src/modules/hisa/createFields.H:29 (discriminator 1)
#15  ? at ??:?
#16  __libc_start_main in "/lib64/libc.so.6"
#17  ? at ??:?
Floating point exception (core dumped)

李东岳

因为你用的别的求解器，不确定是不是求解器的问题，你用出问题的网格，换个求解释发散么？你的第二个例子串行计算

宝丁

是不是你的几何精度问题，壁面网格这么密，画网格的时候如果几何精度太差可能画出来的网格可能会有问题。这里的几何精度既包括建模软件里面的精度，也包括画网格软件转换的精度。如果直接在画网格的软件里面画几何精度应该会好一些，ICEM里面会有调精度的选项，Pointwise不清楚

Foamer24

@宝丁您好！根据您的指导，我查看了Pointwise里面的设置，在Edit—Join Tolerance中有一个设置几何精度的参数，默认值是0.001，我很难确定这个有没有影响，因为以前用fluent计算的时候，这个参数一直是默认的。

Foamer24

@东岳您好！cyclicAMI和renumberMesh之间还存在一些bug，我将cyclicAMI改为cyclic，还是类似的操作，就可以了。需要注意的是：Pointwise导出含周期性边界的网格时，需要先将周期边界设置为patch，然后在openfoam中用createPatchDict将其设为cyclic，就能renumberMesh了。当遇到下面的错误时，提高boundary文件的matchTolerance的值就可以了，不过网格也不能太差。

--> FOAM FATAL ERROR : face 0 and 65 areas do not match by 0.000222148% -- possible face ordering problem 
From function cyclicFvPatch::makeWeights(scalarField& w) const 
in file meshes/fvMesh/fvPatches/derivedFvPatches/cyclicFvPatch/cyclicFvPatch.C at line 62.

CFD中文网

OpenFOAM中cyclicAMI周期边界在有边界层时出现问题