New on DualSPHysics v4.4

edited April 2019 in DualSPHysics v4.4
We are pleased to announce that the new package of DualSPHysics (v4.4) will be released next 11th April 2019 and includes:

Coupling with Project Chrono library
  • Collisions between solids (using restitution and friction coefficient)
  • Realistic frictional behavior with a full Coulomb sliding/sticking/rolling model
  • Large variety of mechanical constraints (joints, hinges and springs)
  • Examples can be found in the subfolder examples/chrono
Coupling with wave propagation models
  • Employing the multi-layered piston wavemaker (with SWASH)
  • Relaxation zone technique for wave and current generation
  • Examples can be found in the subfolder examples/wavecoupling
Open boundary conditions: inlet & outlet
  • A time- and/or space-dependent velocity field can be assigned at an open boundary
  • The density at an open boundary can be either predefined (fixed and hydrostatic options available) or varied according to information retrieved from the adjacent fluid
  • The water elevation at an open boundary can be either predefined (fixed and time-dependent options available) or varied according to information retrieved from the adjacent fluid
  • The buffer areas are built to allow dual inlet/outlet behaviour
  • Examples can be found in the subfolder examples/inletoutlet
Wave generation
  • Automatic generation of single and multiple solitary waves
  • Example can be found in examples/main/16_SolitaryWaves
Novelties on post-processing tools
  • The moment about an axis can be now computed using ComputeForces.exe
More help to produce XML input file
  • Several templates are now available in doc/xml_format
  • New and more complete "XML_v4.4_GUIDE.pdf"
Other changes
  • New definition of domain limits (simulationdomain)
  • Numerical gauge during simulation (gauges)
  • Easy way to restrain the motion (DOF) of floating bodies (examples/others/FtRestrictions)


  • Great!

    Looking forward to it.
  • Can anyone download the DualSPHysics4.4 successfully?
    I can not download from the page of download

    Could anyone who have downloaded successfully send me the code of v4.4?
    Thanks a lot
    My e-mail is

    best regards
  • edited April 2019
    I am confirming that there is an issue accessing the v4.4 package via

    It does appear that 7 lucky people have been able to successfully download the package, though :)

    For the sysadmin:

    After filling out the required fields, passing the Captcha, and submitting the form, the server responds with:

    Array ( [0] => HY000 [1] => 5 [2] => database is locked )

    For what it's worth, I get the same response when attempting to download v4.2, so the problem is not specific to the v4.4 package endpoint.
  • Hi, I want to download the, but I can’t see the Captcha in my computer. Could you send me a link to download it? My email is
  • Try again, it should work better now.
  • edited April 2019
    As of a few minutes ago, I was able to successfully download the package. Thanks, Alex!
  • Hi! Does this build allow restarting a simulation that has Inlet/Outlet boundary conditions?
  • No, restart option has not been finished for Inlet/Outlet and coupling with Chrono
    We have to finish that soon and update again the package....
    But the list of things to be done is too large and we wanted to release the package before Easter.
    We will do our best in the next weeks.

  • edited April 2019
    Hi, I am able to download the package, but in the examples directory 16_WavesSolitary is blank.
  • Thanks for letting us know, we will fix that before tomorrow
  • @Alex
    I have a question about the damping zone in version-4.2 and the relaxation zone in version-4.4.
    Are they the same ? Can they both abosorb the wave(an no reflection boundary)?
    If not , what is the difference between the two zone?


  • edited April 2019
    Many thanks.
    mp4 files are useful and help orient in the essence of the examples.
  • Probably a question for Corrado, is there any benefit in modelling a piston with active wave absorption now that the relaxation zone technique for wave generation has been implemented? Because with the relaxation zone I understand we don't have to worry about re-reflected waves in the domain
  • @Jvaz
    Each generation technique has its own benefits and drawbacks. The advantage of the Relaxation Zone (a part from being also a framework that allows coupling with other models) is the possibility to prescribe accurately the velocity field following the wave theory. So far, relaxation zone does not have a limit related to the water shallowness. In other words, it can used both in deep waters and shallow waters. The particle orbital velocity will be calculated accordingly. However, Relaxation Zone might require more fluid particles than a solution with piston-type wave maker: those particles are the one that initially located in the RZ width and in the damping area before RZ. A piston-type wave maker is an efficient solution and it has been proved to be accurate when used with active wave absorption. A piston does not transfer to the fluid particle the accurate orbital velocity and generates transient waves that will dissipate at a distance equal to 3 times the water depth. But these are drawbacks that can be considered negligible if you use the piston for what the piston is made: intermediate or shallow waters, where the orbits of the fluid particles are flattened and the velocity does not vary so much with the water depth.
    There is then another issue related to wave energy compensation in case of highly reflective structures (reflection about 100%). The RZ shows a sort of warm-up time in the response: it means that a steady regime with accurate wave absorption is achieved only after a few wave cycles. The piston, in stead, does not need this and its solution is accurate from the very 1st wave.
    In a nutshell: unless you have a very specific problem where piston cannot be applied, use piston. If you want to prescribe more accurately the velocity field (even using input from an external model), use RZ.
  • @dong
    The principle behind the damping and the RZ is slightly different but the way it is applied is similar: we act on particle velocities. In both cases the reflection is absorbed, however the damping acs as passive absorber (the velocity is reduced to 0), the RZ cast as active absorber (the velocity is corrected to the target velocity).
  • @iarba27
    Thanks for your reply, it is useful so much for me.
  • @Alex

    I am trying to recompile v4.4 on windows using VS2017 + CUDA 10.0.

    I have modified the file "DualSPHysics4Re.vcxproj" and changed CUDA9.2 to CUDA 10.0.

    After solving a lot of warning, I had:

    fatal error LNK1104: lacking of file "JWaveGen_x64_v141_Debug .lib"

    I noticed in the folder "lib" there is VS2015 with "JWaveGen_x64_v140_Debug .lib".

    So could you offer me "JWaveGen_x64_v141_Debug .lib" based on VS2017?

    Or what I should do to solve it?

  • Please email and he can help you

  • Thanks for the information @Alex

  • Hi
    I just started to learn DualSPHysics.
    I have a question, I would be so grateful if someone could help me.
    As far as I know, these two applications are developed by DualSPHysics:
    -Two-phase liquid-sediment implementation in DualSPHysics
    -Multi-phase Modelling of Violent Hydrodynamics Using Smoothed Particle Hydrodynamics (SPH) on
    Graphics Processing Units (GPUs)

    I wonder whether it is possible to simulate suspension (powders in liquid phase) flow by the current capability of the code or it should be further developed to simulate suspension flow.
    Thank you
  • can I use two GPU cards to run this version?

  • Hi, Alex, I can't see the Captcha. Please send the lastest version package of DualSPHysics (v4.4) to my email 

  • @Asalih3d is there any non-official version where I could run 2 gpu cards?

  • @Asalih3d thanks for prompt response by the way!

  • No problem - I don't think DualSPHysics will provide one, but you could always try to contact them and see - the have other plans of more important stuff, according to the timeline, but it is on the list.

    Kind regards

  • i downloaded update 10 of version 4.4 today and 01_FlowCylinder tutorial for inletoutlet is not running.

    "*** Exception (JSpacePointsMpi::LoadGridPoints) at ..\Source\JSpacePointsMpi.cpp:1319

    Text: No depth data available."

    this message did not appear in the previous update. i checked at github that most recent updates were about inletoutlet and boundcorr bugs... is something missing at this source folder?

  • This problem was solved by updating the package release ( on 5 December 2019.

    Best regards

Sign In or Register to comment.