Yacht in Open Channel Flow
 

I am using Fluent to model a 164-ft yacht in open channel flow.

I have tried running in steady and unsteady flow, as well as trying to switch to steady after running for some time in unsteady flow.

I have predicted values from physical scale model testing.

The grid includes a cylinder encompassing the vessel, which itself is inside a larger domain. There is a grid interface for the "rim" of the cylinder, as well as the "face" of the cylinder. This cylinder is rotated in Gambit to account for the yacht's trim.

The grid includes a layer of prismatic elements around where the free surface is expected. The elements in this region are grown to 2 ft width, 0.25 ft height.

I am using an implicit scheme, VOF, and open channel to model it. I am keeping the boat stationary. I have a boundary layer with 12 layers - yplus is mostly between 60 and 150 in the submerged regions. The boundary layer extends out 1.5 feet from the hull. Some settings are shown below -

Implicit Body Force (on)
Air - primary phase, water - secondary phase
Gravity on
Operating pressure reference point - In the air phase
Specified operating density - 1.225 kg/m^3
Standard k-omega model, transitional flows off, shear flow corrections on
Pressure inlet - open channel. free surface, bottom height, and velocity defined.
Pressure outlet -open channel. free surface, bottom height defined.
Symmetry plane across center of hull (only half the hull is being modeled.)
Top, bottom, side walls are moving walls.
PV Coupling - SIMPLE (PISO doesn't seem to converge.)
Pressure discretization - PRESTO!
Momentum - 2nd order upwind
Volume Fraction - Modified HRIC
Turbulent Kinetic Energy - 1st order upwind
Specific Dissipation Rate - 1st order upwind
Pressure URF - 0.5
Density URF - 1
Body Forces URF - 1
Momentum URF - 0.2
Volume Fraction URF - 0.2
Turbulent Kinetic Energy URF - 0.5
Specific Dissipation Rate URF - 0.5
Turbulent Viscosity URF - 0.5
Initialization process - (rpsetvar 'patch/vof? #t) , initialize with velocity set, and water volume fraction 0, Mark the water region, Patch the water region with water VF = 1.

Running in steady starts to converge, then diverges. The free surface doesn't look reasonable, and the drag on the boat is less than half of what is expected.

I ran in unsteady with time steps of 0.000125, 0.0004, 0.00125, 0.004, 0.0125, 0.04, 0.125, 0.4, 1 second, about 15 time steps each. They converge, but the drag on the boat is about twice as big as it should be. If I continue running time steps, the drag starts to decrease very slowly - but it should have already reached a "steady" state.

I'm not sure what else to try to get this to work. Any help would be much appreciated!!!

To those who say Hollywood is filled with satan worshippers and sodomites, heres your post. Aleister crowley made this one. That’s great to hear! Please let us know what you think of it. I am greatly enjoying it, myself. Thanks for the update.


SeaDream Yacht Club