Twenty-Fourth Symposium on Naval Hydrodynamics (2003) / Chapter Skim
Currently Skimming:
Validation of the Flow Around a Turning Submarine
Validation of the Flow Around a Turning Submarine
Pages 669-681
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 669... ...
This is because RAE can be treated as a time independent problem in a steady rotating coordinate system thus avoiding complications by time dependency and other oscillatory flows. Yet RAE maintains most of the major flow physics unique to a maneuvering submarine such as vertical flow interactions with hull boundary layer and flow separations.
|
From page 670... ...
This is expected since it can be proved that turbulence models must be form invariant under arbitrary translational acceleration of the reference frame and should only be affected by rotation through the mean vorticity tensor[Si. The rotational effect is thus introduced explicitly only in non-linear models through the mean vorticity tensor defined in (10)
|
From page 671... ...
IFLOW is intended to be a general- purpose production code for solving 2D, 3D, steady and unsteady problems. The code is highly modular in structure so different turbulence models and newly developed numerical schemes such as higher order schemes can be easily implemented.
|
From page 672... ...
A 2-level multigrid consisting of the medium and coarse grids is run for about 50 cycles and then the solution is extrapolated to the fine grid to start the 3-level multigrid computations. In general, a solution adequate for engineering applications can be achieved in 300-500 multigrid cycles.
|
From page 673... ...
The lower bound of ~ for the k—~ model is derived in a similar manner and is given by r ~ > C~f~ky/-Po DESCRIPTION OF EXPERIMENT (28) A Rotating arm experiment was performed on an unclassified generic submarine model called ONR BODY-1 in the Rotating Arm Basin at NSWCCD as reported in Reference [11.
|
From page 674... ...
The fine, medium and coarse grids having a total of 2.89, 0.36 and 0.075 million grid cells, respectively. For computer runs in this paper, the non-dimensional normal distance to the wall from the first grid cell center, y +, is approximately 1 for the fine grid, 3 for the medium grid and 6 for the coarse grid.
|
From page 675... ...
Convergence history of force Z' t -- -- 1 1 L -6.0 ~ ~ 100 of · test data coarse grid 0 - - medium grid fine grid 1 1 1 1 1 1 1 1 1 1 1 ~ I I I I , . , , 200 300 400 500 multigrid cycles the time dependency ofthe flow at such a high value of r ' = 0.53.
|
From page 676... ...
fine grid _ 1 ~ 1 ~ 1 1 1 1 1 ~ 1 1 1 1 1_ ~ 1 1 1 1 400 500 second row and the standard deviations in percentage deviations from the mean values are presented in the third row. The predicted values with the coarse grid, medium grid and fine grid using the realizable k—~ turbulence model are then presented.
|
From page 677... ...
For a turbulence model to perform well in the prediction of the flow about a turning submarine it is essential that it can predict accurately the force and moment of a bare hull at incidence. One of the most important flow physics about a turning submarine is the interaction between the vertical flow shed from the sail and the cross flow on the hull boundary layer caused by rotation.
|
From page 678... ...
The predicted crossbow of the vortex shed by the sail at L, = 0~34 compared well with the PIV laser sheet. The interaction between the vortex shed by the sail and the hull boundary layer results in a downward force in the aft hull and consequently a pitch-up of the submarine nose.
|
From page 679... ...
It is also clear from this investigation that a grid with 3x 1 o6 cells is far from sufficient for accurate prediction. The majority of the grid cells are distributed near the wall because conventional turbulence models require the first y + be approximately 1.
|
From page 680... ...
4. Shih, Tsan-Hsing, Jiang Zhu and John Lumley, "A New Reynolds Stress Algebraic Equation model", NASA Technical Memorandum 106644, ICOMP-94 -15; CMOTT-94-08, August 1994.
|
From page 681... ...
We agree with the discusser that a far greater number of grid cells is needed. Our next attempt will still be based on the structured grid but with more sophisticated numerical techniques including local refinement, higherorder spatial discretization schemes and a new wall function method for the turbulence models.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.