[Mojtabam]

Dear all
I have been trying to rotate a high speed craft about its center of gravity in order to find its hydrodynamic coefficients. In attached you can find the image of this vessel in CFX environment. In fact, for deriving the hydrodynamic coefficients of a vessel two sets of simulation are conducted, pure heave( motion in z direction) and pure pitch( rotation about y axis). Pure heave simulation was carried out successfully by defining an appropriate expression. However, I have encountered some difficulties in pure pitch simulation.I need some useful expressions which can be used to do the pure rotation about an perpendicular axis on a rigid body( in this case a vessel).
I would be so grateful to take some helpful advice from you,
Thank You
Mojtaba

[ghorrocks]

If I was doing this I would generate a mesh at all the different rotations. I think there is a new morphing thing which might be able to help here but I am not familiar with it.

[Mojtabam]

Dear Mr. Glenn Horrocks
Thank you for your reply. Actually I conducted this simulation by defining some appropriate Expressions which considered the new position of mesh instead of displacement.
New X=
if(t>Timeoffset ,((x-X0)-Total Mesh Displacement X)*cos(0.1 [rad]* cos(1 [s^-1] * t + Offphase))+((y-Y0)-Total Mesh Displacement Y)*sin(0.1 [rad]* cos(1 [s^-1] * t + Offphase))+X0 , 0 [m])
New Y=
if(t>Timeoffset ,((y-Y0)-Total Mesh Displacement Y) + Y0, 0 [m])
New Z=
if(t>Timeoffset ,-((x-X0)-Total Mesh Displacement X)*sin(0.1 [rad]* cos(1 [s^-1] * t + Offphase))+((z-Z0)-Total Mesh Displacement Z)*cos(0.1 [rad]* cos(1 [s^-1] * t + Offphase))+Z0, 0 [m])

[ghorrocks]

To do this you need a transient simulation with moving mesh. This is a very inefficient way of doing it and will take a very long time. A series of steady state runs as I described will be far more efficient.

[Mojtabam]

However I need to do my simulations in transient way. because I need time history of the forces acting on the vessel. I dont think that steady state simulation would be helpful for me.

[ghorrocks]

I thought you were deriving the motion coefficients. Can't they be done steady state?

[Mojtabam]

Motion coefficients are derived from the forces equations acting on the vessel. these forces must be a function of time. so it is essential to do the simulation in a transient way.

[ghorrocks]

I will take your word for it.

How are you modelling the vessel?

As for your original question - wouldn't a transformation matrix do what you want (http://en.wikipedia.org/wiki/Transformation_matrix)?

[Mojtabam]

the vessel is considered a rigid body with no degree of freedom and no slip wall for its surface. For modeling the free surface the same expressions as for Buoy which you can find it in CFX tutorial.Transformation matrix is used for changing the coordinate system but here we have time.

[ghorrocks]

? Then you make the transform matrix a function of time. Isn't that obvious?

[Mojtabam]

Dear Mr G. Horroks
I must admit that I did not totally understand your suggestion. I would be so grateful to you if you could give me a vivid explanation of your suggestion.

[ghorrocks]

I will be a little more vivid: For a simple 2D rotation you get....



make theta=a*sin(t) and it will rock side to side as a function of time. If you want a more complex transformation look at the wikipedia reference I quoted.

In your post #3 you have a function which looks similar, but has several extra parameters. Is there something wrong with these functions?

[Mojtabam]

I think it would be a good idea to use transformation matrices. I am going to make use of this idea, and I think it would save more time for me. The Expression in post #3 is actually like the following:
New X=
if(t>Timeoffset ,((x-X0)-Total Mesh Displacement X)*cos(alpha)+((y-Y0)-Total Mesh Displacement Y)*sin(alpha)+X0 , 0 [m])
Timeoffset is set by the user, for example I set this variable, 0.
alpha= 0.1 [rad]* cos(1 [s^-1] * t + Offphase)
the vessel will be rotating with an amplitude of motion equal to 0.1 rad and frequency equal to 1 s^-1.