@strive_DUT 谢谢

我最近也碰到这个问题，单核计算时对某个特定的patch上的面心可以赋值，但并行分块之后也是运行不了。

Hi https://cfd-china.com/user/tidedrinker

我在使用of2306也出现类似的情况，我的波频是1.3 而且不同网格密度的相位差和衰减率有明显差异

但是我在使用最密网格（单位波高内20个网格）时，总是卡在67 迭代不出来 最终计算发散

Courant Number mean: 1.18882076445e-06 max: 7.50119780436 Interface Courant Number mean: 3.00160678714e-08 max: 3.95428576669 deltaT = 6.18380398924e-152 Time = 67.132 PIMPLE: iteration 1 Updating StokesI wave model for patch inlet smoothSolver: Solving for alpha.water, Initial residual = 5.38442986655e-09, Final residual = 5.38442986655e-09, No Iterations 0 Phase-1 volume fraction = 0.666618301469 Min(alpha.water) = -2.29205431887e-10 Max(alpha.water) = 1.00000001308 MULES: Correcting alpha.water MULES: Correcting alpha.water Phase-1 volume fraction = 0.666618301469 Min(alpha.water) = -2.29205431887e-10 Max(alpha.water) = 1.00000001308 Updating shallowWaterAbsorption wave model for patch outlet DICPBiCGStab: Solving for p_rgh, Initial residual = 0.844109801632, Final residual = 0.00558442869736, No Iterations 20 DICPBiCGStab: Solving for p_rgh, Initial residual = 0.742328538632, Final residual = 0.00717435702295, No Iterations 26 time step continuity errors : sum local = 1.13821305653e-09, global = -5.40663033957e-10, cumulative = 2.0897431155e-05 DICPBiCGStab: Solving for p_rgh, Initial residual = 0.823953322827, Final residual = 0.00800680485496, No Iterations 25 DICPBiCGStab: Solving for p_rgh, Initial residual = 0.867377012998, Final residual = 9.38111948679e-10, No Iterations 389 time step continuity errors : sum local = 9.20755023008e-16, global = 7.16963646939e-16, cumulative = 2.08974311557e-05 smoothSolver: Solving for omega, Initial residual = 7.31631987399e-08, Final residual = 4.57762864183e-11, No Iterations 1 smoothSolver: Solving for k, Initial residual = 4.39226774226e-07, Final residual = 2.03583449134e-10, No Iterations 1 ExecutionTime = 29318.96 s ClockTime = 29376 s Courant Number mean: 1.18884127695e-06 max: 7.50169136387 Interface Courant Number mean: 3.0019195664e-08 max: 3.95459850554 deltaT = 8.24321301597e-153 Time = 67.132 PIMPLE: iteration 1 Updating StokesI wave model for patch inlet smoothSolver: Solving for alpha.water, Initial residual = 5.38393271075e-09, Final residual = 5.38393271075e-09, No Iterations 0 Phase-1 volume fraction = 0.666618301469 Min(alpha.water) = -2.29205431887e-10 Max(alpha.water) = 1.00000001308 MULES: Correcting alpha.water MULES: Correcting alpha.water Phase-1 volume fraction = 0.666618301469 Min(alpha.water) = -2.29205431887e-10 Max(alpha.water) = 1.00000001308 Updating shallowWaterAbsorption wave model for patch outlet DICPBiCGStab: Solving for p_rgh, Initial residual = 0.844175076514, Final residual = 0.00539645720402, No Iterations 20 DICPBiCGStab: Solving for p_rgh, Initial residual = 0.742024848316, Final residual = 0.00733560856723, No Iterations 27 time step continuity errors : sum local = 1.16277362766e-09, global = -5.78716682763e-10, cumulative = 2.08968524391e-05 DICPBiCGStab: Solving for p_rgh, Initial residual = 0.824114713535, Final residual = 0.00803020015335, No Iterations 25 DICPBiCGStab: Solving for p_rgh, Initial residual = 0.867366339743, Final residual = 9.11722968482e-10, No Iterations 373 time step continuity errors : sum local = 8.95106772158e-16, global = -5.97636368108e-16, cumulative = 2.08968524385e-05 smoothSolver: Solving for omega, Initial residual = 7.30573956466e-08, Final residual = 4.54407875238e-11, No Iterations 1 smoothSolver: Solving for k, Initial residual = 4.38607733913e-07, Final residual = 2.03320386843e-10, No Iterations 1 ExecutionTime = 29358.38 s ClockTime = 29416 s Courant Number mean: 1.18886725019e-06 max: 7.50212848379 Interface Courant Number mean: 3.00241095746e-08 max: 3.95487860459 deltaT = 1.09878323649e-153 Time = 67.132 PIMPLE: iteration 1 Updating StokesI wave model for patch inlet smoothSolver: Solving for alpha.water, Initial residual = 5.38400281089e-09, Final residual = 5.38400281089e-09, No Iterations 0 Phase-1 volume fraction = 0.666618301469 Min(alpha.water) = -2.29205431887e-10 Max(alpha.water) = 1.00000001308 MULES: Correcting alpha.water MULES: Correcting alpha.water Phase-1 volume fraction = 0.666618301469 Min(alpha.water) = -2.29205431887e-10 Max(alpha.water) = 1.00000001308 Updating shallowWaterAbsorption wave model for patch outlet -------------------------------------------------------------------------- Primary job terminated normally, but 1 process returned a non-zero exit code. Per user-direction, the job has been aborted. -------------------------------------------------------------------------- -------------------------------------------------------------------------- mpirun detected that one or more processes exited with non-zero status, thus causing the job to be terminated. The first process to do so was: Process name: [[8254,1],7] Exit code: 145 --------------------------------------------------------------------------

在粗网格和中等网格却计算很稳定。根据https://cfd-china.com/topic/5462/mpirun-detected-that-one-or-more-processes-exited-with-non-zero-status，尝试修改了fvSolution，但依旧迭代不出来，请问您有什么解决办法吗

https://link.springer.com/article/10.1007/s11440-023-01982-6 可以看下这个文章

我参考了phaseScalarTransport.C的方法，人为地乘以了alpha来区分界面

这个也不是特别的稳定。

@李东岳 好的 感谢东岳老师

@dxl 我印象中拉格朗日这边OF5以后发生了一些变化，不知道你用的什么版本。
但是归根到底就是在拉格朗日那边有一个判断。后续版本好像转为p.fraction了
如果dt太小，这拉格朗日的计算就跳过了

https://github.com/OpenFOAM/OpenFOAM-8/blob/master/src/lagrangian/intermediate/parcels/Templates/KinematicParcel/KinematicParcel.C
line 334-348
// Avoid problems with extremely small timesteps
if (dt > rootVSmall)
{
// Update cell based properties
p.setCellValues(cloud, ttd);

p.calcDispersion(cloud, ttd, dt); if (cloud.solution().cellValueSourceCorrection()) { p.cellValueSourceCorrection(cloud, ttd, dt); } p.calc(cloud, ttd, dt); }

好久没碰了，特别细节忘记了。

@lyc 不好意思，我不知道答案

@李东岳 平时工作一半的时间做CAD制图&检图，少数时间用到FEM和CFD（舾装就是涉及的比较多，比较杂）。去年年初开始接触openFoam一直磕磕绊绊，下半年从您这里拿到64核神器开始有点进展。:146:

@李东岳 是的李老师，都是涡量输运方程，但不是一个实验室哈哈哈:chouchou:

你这样写可以没问题

一阶时间格式比如Euler这种和二阶backward会产生很大的差异吗？

不会

f375c892-9e6a-4a68-8b65-1f29a17d4b90-图片.png
目前有个需求是需要在计算域出口降低y方向网格密度，但直接改block的y反向网格数目会生成非结构化网格，显然行不通，之前想过用refinemesh等局部加密的功能反向加密，但了解了发现只能加密不能系数，友友们有了解过怎么稀疏网格吗？或者说blockmesh能不能生成非结构化网格？

已经找到解答：https://cfd-china.com/topic/1553/并行-获取某个量达到最大值的这个网格中的其它值?_=1698117342935

@李东岳 感谢东岳老师！

@学流体的小明 谢谢，我去看看

提示信息如下：
...
rt -lm -o /myelasticThermalSolidFoam
/usr/bin/ld: cannot open output file /myelasticThermalSolidFoam: 权限不够
collect2: error: ld returned 1 exit status
make: *** [/home/zhangec/foam/foam-extend-5.0/wmake/Makefile:160：/myelasticThermalSolidFoam] 错误 1
请教各位老师同学怎么解决？

@李东岳 哪想获得稀疏网格就没有办法了嘛?比如说我需要一段粗糙网格去增加耗散

@李东岳 李老师好像对这个比较懂哈哈

以及对于入口气态氧化剂超声速射入燃烧室，入口液态燃料低速进入燃烧室进行非预混燃烧，最后速度为低马赫数的用sprayFoam进行燃烧仿真合适吗
请各位给小白点建议

同问，我也出现了这个问题，不知道该不该去调整