Wave Attenuation in Flume

Hi There,

I'm running an experiment studying wave attenuation through oyster growing cages. Part of the study is using a SPH simulation to look at wave attenuation. I've created a "flume" similar to the examples. Immediately, what I'm seeing is too much wave attenuation. I investigated further, and am seeing that even in an empty flume, I have wave heights decreasing across the domain. I'm unsure of the cause of this. I have a sloped beach at the end of my domain, but I'm not seeing major distortion from reflection. I've attached a paraview snapshot of my domain below.

I'm running at a dpi of .02m, and using a piston wavemaker with irregular waves. I'm quantifying attenuation using a wave spectrum at the first and last gauges, and seeing about 70% of wave energy dissipated in an empty flume. I'm using the MDBC. Does anyone have any suggestions?

TLDR; I expect to see little to no change in waves across flume, but am seeing waves decrease in energy.


  • I realize my issue is the same at the one described here:

    So i will follow these suggestions and see if there is improvement.

  • Awesome!

    Let us know, then we can help hopefully

    I think AWAS would do wonders here. Some wave attenuation is properly to be expected still.

    Kind regards

  • Thank you!

    Right now I have implemented all suggestions but AWAS and am seeing about 90% attenuation. This is much better than before. I'm now running one with AWAS and will see those results in the morning

  • Let us start with a reminder: SPH method is not designed to study wave propagation or wave attenuation, it is a method to study wave-structure interaction.

    IN addition, we can propagate waves... yes... but standard SPH formulation can only propagate waves without wave decay for 2-3 wavelenghts maximum... using more sophisticated formulations such as delta-SPH can allow you to propagate waves better and longer.... and of course, other approaches as using Riemann Solver implementations can avoid this problem.

    In the case of DualSPHysics using a proper viscosity treatment with DDT Fourtakas can help to propage waves perfectly for more than 3 wavelenghts


  • I can confirm that the suggestions posted in the thread below do in fact work. I'm running waves in a flume about 45m long and see very little attenuation. I don't use mDBC or AWAS in my simulation. In addition, be sure your particle spacing is about 1/10 the wave amplitude.

  • In addition, using coefh=1.2 or higher will also help to propagate waves for 3 wavelengths at least


Sign In or Register to comment.