Simulate sloshing using SPH method

[Ganesh Persaud]

How can I simulate sloshing using the smooth particle hydrodynamic method in ABAQUS? Can someone ref me to a sample procedure utlizing time history data. The problem I’m having is how do I input the time history data and specify the boundary conditions to accurately simulate sloshing.

 

[FEA way]

Applying boundary conditions with amplitudes to force proper motion of the tank is the easiest part since it’s done in the same way as in other kinds of analyses. What’s more challenging is the creation of particle elements since they are not normally supported by Abaqus/CAE. One way is to create point masses and edit input file to convert them to PC3D elements. Another option is to use finite element conversion to SPH particles feature (Mesh —> Element Type —> Conversion to particles).

 

[Ganesh Persaud]

Creation of the particles is not a problem. I’ve meshed the fluid component and yes I’ve applied a very fine mesh and used the SPH conversion.

I ran an analysis of the tank using an acceleration, got the simulation. Model run time was long.

My only problem is converting the time history, xyz data. What do I input it as, and then apply three boundary condition to one node of the tank for each xyz, direction??

There’s limited info on this with abaqus.

 

[FEA way]

What kind of time history data do you have ? Time vs what ?

You may have to prepare the data in spreadsheet, divide it into three spatial components and define separate amplitudes that will be used for boundary conditions in Abaqus. You will need separate BCs applied to different degrees of freedom.

 

[Ganesh Persaud]

I have time history data in time (sec) vs Acceleration (g) for xyz directions.

The below image is what I've done but there is no actual ref to what I've done to say that it's correct. What's your take on it

 

[FEA way]

It will be sufficient to define acceleration boundary conditions in 3 directions with references to proper amplitudes.