[Prenaud]

Hello everyone,

I am currently working on OFv1806 and interFoam to simulate a (full scale) boat in calm waters with dynamic trim and sinkage. I am able to reproduce the simulation from this post for a planing hull Planing boat simulation with InterDyMFoam and force ramp for make it stable

, but I can not get a good result for displacement hull.

Indeed, the bow sinks in the water until the crash of the simulation. I do not understand why. Whether taking the parameters of the DTCHull or simulation with planing hull (boundary conditions, fvSchemes, fvSchemes ...), I always get the same trend.

I tried initializing with dynamic mesh (LTS), adding a ramp force, modifying accelerationRelaxation, passing from rigidibodyMotion to sixDofRigidBodyMotion without success. In DynamicMeshDict, I put the centerOfMass as the center of gravity, momentOfInertia and the mass of the half hull (because of the symmetry).

Some data :
L = 6.4 m
Draft = 0.484 m
Displacement 2.707 m3
U = 2.57 m/s
k = 0.0003566 m2/s2
nut = 5e-07 m2/s
omega = 495 /s

Time step = 0.001 and Courant number max = 0.5 before the crash

Mesh = 0.5M cells with layers
Has anyone ever faced this problem?

Kind regards

boat_sinks.jpg

mesh.jpg

dynamicMeshDict.zip

fvSchemes.zip

fvSolution.zip

[Yann]

Hello Paul,



The first picture shows a quite distorted mesh with cells which could cause a crash. Have you tried increasing the outerDistance parameter? It could help to smooth the mesh deformation.


Yann

[Prenaud]

Hello Yann,

Yes I have already tried to increase this parameter. And it was actually the cause of the crash in another simulation where the (positive) trim could not increase more because of the ¨compression¨ of the mesh.

Paul

[Prenaud]

My best simulation on that hull has been quiet stable but the trim should be positive and not negative. On all the other hulls, the simulation has the same behavior as before with the bow that sinks.

Paul

trim_monitoring.png

[Artur]

Hi,


Your Froude number of 0.32 is quite high for a displacement hull, so the boat will probably enter semi-displacement mode, is that correct?


If you know what the trim should be, have you tried running a static case with the hull at the correct orientation (known trim and sinkage), and confirming that the pitching moment is zero and the total vertical force equals m*g?


My suspicion is that with a rather coarse mesh of 0.5 million cells, you might simply be under-predicting the dominant flow features, such as wave pattern and separation around the stern, which in turn will affect your pressure distribution and hence the trim. You said you got good results for a planing hull with a similar set up - that should be quite a lot easier to predict since the flow around the stern and chines will separate cleanly, unlike on the boat in question. At least that would be my expectation.


Happy foaming,


Artur

[Prenaud]

Hello,

Actually, I don't have the exact value of the trim and not at all that of the sinkage so I have not tried this option.

I take note of your remark and redo with a finer mesh.

Thank you for these tips.

Paul

[Artur]

Right, I see. Is that a hull for which you have experimental data to compare against? If not, maybe it'd be worth trying one of the NPL hull series ones first, since they're also semi-displacement craft? I've got some data here, but there's a lot more on the internet:
https://www.dropbox.com/sh/5tz4iw4ff...bPYJNul7a?dl=0


Alternatively, there are some free to sink and trim data for the well-known KCS, albeit at lower Froude numbers:
https://t2015.nmri.go.jp/kcs.html
The Duisburg Test Case from the tutorials is also a good one but I have less first-hand experience with it.


Good luck,


A

[Prenaud]

Hello,


For now, I try to have convergent simulations on all types of hull ranging from 5 to 50m, then I would compare with experimental results or with other CFD study. For this boat, I just have estimates of the trim so yes it is not suitable for comparison ...

Thank you for the link. I would probably use these hulls to focus on the accuracy of the results.

Paul

[vava10]

Hey,

I am simulating a ship. Due to symmetry I am only working with half of the geometry. for dynamicMeshDict I have following doubts

1. centre of mass - the centre of half or full geometry
2. moment of inertia -moment of inertia of half geometry or (moment of inertia of full geometry)/2


Thanks in advance
vava10

[Artur]

Hi,
Unless you're using some fancy ship-related add-ons, halve everything because what matters are the properties of the object being resolved by the grid.
Cheers,
A