[rangure]

hi,
I'm trying to use openfoam to simulate a wave flume with a slop. and to generate a wave, a flap wavemaker on the left-hand side is implemented with dynamic mesh movement. Now since the displacementLaplacian solver acts on the whole domain, the slop will have non-physical movement with the mesh movement. To solve this problem I'm using frozenPointsZone to freeze the points on the right. For example, suppose the total length of the flume is L and the slope starts at x from the left wall. Then the mesh in [0,x] will move according to the diffusivity solver and the mesh in [x,L] will stay fixed. Now the problem is that the inverseDistance diffusivity is using L as its base(1/L). This will cause problems if I have a large L and a relatively small x, as the first column of blocks on the left of x will be significantly smaller than other blocks. Is there any way I could use x as the base of the inverse calculation? or any of the diffusivity openfoam provided can mitigate this problem. here's my current setup and inlet is the left wall.

dynamicFvMesh      dynamicMotionSolverFvMesh;
motionSolverLibs   ("libfvMotionSolvers.so");

solver            displacementLaplacian;

displacementLaplacianCoeffs
{
    diffusivity quadratic inverseDistance (inlet);
    frozenPointsZone fixedZone;
}