Learning Forum

[Abdul_4]

Hi all,

I have been working on floating body analysis and created a 3D wave tank with a floating body in FLUENT. The waves are generated properly for empty tank and waves are started from inlet. The problem encountered with a floating body, the waves are starting from the structure instead of inlet and simulation fails with floating point exception. I have provided proper boundary conditons and tried simulations by changing the mesh size, scaled flumes, scaled structres and solving methods, but the simulations are failed with floating point exception after few iterations. 

 

kindly let me know the solution for this issue, if anyone are aware of this.

 

Thank you.

 

Please check the pictures for the reference.

[Rob]

The free surface looks very diffuse. Best guess is you need to check the mass & initial position of the floating object. Ie should be be where it is? You may find you need a smaller time step to get going as the initial bobbing around could be quite rapid. 

[Abdul_4]

Thank you for your response dear Rob

Even though it is a floating body, the postion of the body is fixed in FLUENT and it defined as wall. The free surface diffusion can be reduced with smaller mesh size which  I have tried. My main issue for the present model was, the flow is starting at the location of structure instead of inlet. The empty flume works fine.

Please let me know any further suggestions for this issue.

 

Thank you.

[Rob]

If the structure is fixed then the waves may reflect but shouldn't "start" from there. 

[Abdul_4]

Thank you for the response dear Rob

I have seen few tutorials and the same way provided the boundary conditons as per my study requirements. The flow is not converging and starting from the structure. I have tried for another structure (box shaped) which is bottom fixed to sea bed with same boundary conditons and its working well. I didn't understand the exact problem with floating model.

Please let me know any further suggestions to solve this issue.

 

Thank you.

[Rob]

How are you starting waves from the structure if it's fixed? 

[Abdul_4]

Dear Rob

I am also having the same doubt. I have given velocity at inlet and the wave is starting from the structure.

This is the issue i am facing.

 

Please let me know if know the relavent solution.

 

Thank you.

[Rob]

OK, so the wave isn't starting from the structure, the model instability is. Review the mesh around the structure. Is it well resolved? What cell quality have you got? Did you use inflation? 

[Abdul_4]

Dear Rob,

I have provided the 6cm wave height (third order Stoke's theory) at the inlet, but the wave is not starting from the inlet. 

 

Meshing details:-

I have given face sizing for the structure and didn't use inflation.

Nodes:- 437302

Elements:- 2330736

Element Quality: 0.19651 Min- 0.99992 Max

 Aspect ratio of mesh: 9.9957 Max -- 1.1656 Min

 

Thank you.

[Rob]

And the cell size near the structure? How did you initialise the solution? 

[Abdul_4]

 

The cell size near the structure is about 35mm and the fluid is also 35mm. I tried to get the aspect ratio below 10, to get a convergence.

I have tried for both Hybrid and Standard Initialisation – Flat type.

Thank you. 

 

[Abdul_4]

Dear Rob

Please check this image once for the reference.

Thank you.

[Rob]

You have a wave of 60mm and a cell of 35mm?  That's not going to give a very good result.

The wave you're showing is down to the initial condition: replot with node values off. If you want the free surface to be completely flat you may need to split the domain to get the patch in exactly the correct location. 

[Abdul_4]

Dear Rob

Thank you for the response.

I have kept simulations as you suggested and I think it's working.

 

Thank you.