[Chubo]

Dear Foamers,

I am new in OpenFOAM simulation. Recently, I am building a numerical wave flume by using OLAFOAM library, to simulate the motion of a wave energy converter (olaDyMFoam). The flume works well. But when I add a function to the controlDict to run-time sample alpha.water at fixed points, the solver "olaDyMFoam" only runs 9.7 sec and then stop updating the log file (log.olaDyMFoam). The odd thing is, this job will be still running until timeout without any errors.

I use OpenFOAM-3.0.1 and my job was calculated on a cluster. The controlDict setting is listed below. Many thanks for your help.

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  3.0.1                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     3.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      controlDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

application     olaDyMFoam;

startFrom       latestTime;

startTime       0;

stopAt          endTime;

endTime         240;

deltaT          0.1;

writeControl    adjustableRunTime;

writeInterval   0.05;

purgeWrite      0;

writeFormat     ascii;

writePrecision  12;

writeCompression uncompressed;

timeFormat      general;

timePrecision   6;

runTimeModifiable yes;

adjustTimeStep  yes;

maxCo           .5;
maxAlphaCo      .5;
maxDeltaT       0.02; 

functions
{
    waveElevation
    {
        type                   sets;
        functionObjectLibs     ("libsampling.so");
        setFormat              raw;
        interpolationScheme    cellPoint;
        outputControl          outputTime; // Don't use timeStep
        outputInterval 1;
        sets
        (
            Probe_50m
            {
                type        uniform;
                axis        z;
                start       (50 0 2);
                end         (50 0 -2);
                nPoints     401;
            }
            Probe_100m
            {
                type        uniform;
                axis        z;
                start       (100 0 2);
                end         (100 0 -2);
                nPoints     401;
            }

        );
        fields
        (
           alpha.water
        );
    }
};


libs
(
    "libwaveGeneration.so"
    "libwaveAbsorption.so"
    "libsampling.so"
);

[Phicau]

Hi Chubo,

I have faced this problem too.

This is caused by a limitation of OpenFOAM sampling procedure. From what I remember, as the time and mesh deformation progress, the initial sampling points that were inside a given cell will belong to another one (or even to a face, which is less likely, but has also happened to me), and sampling will fail.

I suggest that you open a bug in the OpenFOAM system to see if they can do something to update the points at runtime for 'DyM' solvers.

https://bugs.openfoam.org/rules.php

Best,

Pablo

[Chubo]

Hi Pablo，

Thank you for your information.

Not sure the problem will be solved if I change the interpolationScheme from 'cellPoint' to 'cellPointFace'. I will let you know the result.

Regards
Chubo