Seems like a conflict for reality - deformation means some parts must have different velocities. Do you mean constant speed while allowing the direction of motion at each point to change? or by “entire plate”, do you mean its center of mass or the point where it’s attached to something?
Other software has RBE3 which would allow you to enforce a constant average velocity while keeping all the nodes free to move relative to each other. CCX doesn’t have that but depending on what you’re doing, you might be able to approximate it with springs or write constraint equations to set the mean displacement equal to the displacement of a single node which has the prescribed displacement boundary condition.
Or maybe you can model it in the plate’s frame of reference if you know what the external forces will be? Perhaps with gravity as a way to provide reaction force distributed over the plate?
Perhaps you can explain a bit about what you’re trying to model?
Thank you for your suggestion. The experimental background of the case I want to compare is that the plate enters the water at a constant speed exerted the external force of the experimental device. But for my simulation, this external force is variable and unknown. I only know the constant speed. The case in this paper is what I want to simulate. The author seems to have used ABAQUS to achieve the constant speed motion of the plate:Redirecting
I have an idea. I can only apply displacement to the nodes on the left and right sides of the plate, without applying displacement to the nodes on other surfaces. In this way, most nodes can be deformed and can also maintain constant speed. I am planning to try it. I wonder if it is feasible? What do you think?
Looks close to what they did except they made the whole perimeter rigid, not just two edges, and tried it as both clamped and pinned:
“The experimental plate is attached to a thicker aluminum frame that is
attached to the test rig. To compare with the experiments, the FEM model
applies either clamped or pinned boundary conditions where the plate meets the inner edge of the frame.”
So it looks like you should perhaps have not only prescribed displacement in the direction of motion, but also zero displacement (and optionally rotation) BCs in the other two directions.
On the other hand, elsewhere, they used a completely rigid plate:
“an approximate velocity boundary condition approach, where the velocity is applied on the fluid domain at the undeformed position of the structure, and the mesh is not deformed.”
trust Newton’s first law, if you apply initial conditions with constant speed it will keep that speed as a free body until it collides. See following example using gravity: _imp2.inp - Google Drive. Credits go to calculixforwin.
In the linked paper, they’re forcing the plate to maintain constant velocity even after it hits the water, which you wouldn’t get with just initial conditions.
I suppose a small panel would be forced at constant velocity by the inertia of the rest of the aircraft. But this is all just an idealized scenario with a lab measurement anyway so whatever their reason, that’s how they did it.
Yes, because in the experimental scenario, the speed is constant in most cases, which makes it easier to do experiments.
I failed again. I gave the node displacement boundary conditions on the plate around the plate, but unfortunately, only the nodes on the boundary are moving, and the other nodes still do not move. Then bending occurs, as shown in the figure. Do you have any other suggestions?
If I apply an initial velocity to all nodes, it will work fine. But in that case, it is hard to know whether the entire plate will move at a constant velocity or initial velocity, even with constant displacements on all four edges.
Thank you very much for the information you provided. But as shown in this paper. The device is always connected to the plate, and after accelerating to a certain speed, it moves forward entering the water at a constant speed. In the simulation, we only need to give a constant initial speed. For details, you can look at the information:
Iafrati A. Experimental investigation of the water entry of a rectangular plate at high horizontal velocity. Journal of Fluid Mechanics . 2016;799:637-672. doi:10.1017/jfm.2016.374
Iafrati, A., Grizzi, S., Siemann, M.H. and Montañés, L.B., 2015. High-speed ditching of a flat plate: Experimental data and uncertainty assessment. Journal of Fluids and Structures , 55 , pp.501-525.Redirecting
that’s what I said, plate isn’t forced to maintain constant velocity AFTER it hits the water. So you only need to apply initial conditions type velocity. When I said “releases” I meant from the catapult strings, but is still guided BUT free.
Hello, after my test, it is not feasible to rely on the initial velocity in the simulation. After the release, the plate bounces up immediately when it encounters water, and will not be immersed in water at all. Therefore, I concluded that the experiment was also carried out under constrained conditions.
