Nuts

CFD Applications in Ship Design

Optimization

1

Khairul Hassan

Doctoral student in Department of Maritime Engineering

Graduate School of Engineering, Kyushu University, Japan

Maurice F. White

Professor of Marine Engineering

Department of Marine Technology

Norwegian University of Science and Technology (NTNU) Norway

Cosmin Ciortan, PhD, Consultant

Dept. of Ship Hydrodynamics, Det Norske Veritas (DNV),

Oslo, Norway

2 Introduction

Brief description of the CFD procedure

CFD application

CFD application in ship design optimization

CFD application in drag analysis for different wind directions

Limitations of the CFD simulation

Conclusion

3During design optimization the important considerations

ship capacity and Ship stability

Ship Hull

Hydrodynamic

resistance

Aerodynamic

resistance

CFD simulation in Ship design optimization

Ship design optimization

Dimensions optimization

Shape optimization

CFD simulation can be used for

both of the

optimizations

4For wind resistance simulations, only the part above the waterline is

considered

Ship Hull

Geometry of the problem

Principle particulars

Length water line, LWL=221.65m

Breadth=32.2m

Depth=18.5m

Draught=10.78m

Block coefficient, CB=0.674

Deadweight, DWT=40900tonnes

Cargo capacity: 2800TEU containers;

Design speed: 23 knots

Boundary conditions and simulation conditions

Simulation Space 3 dimensional

Motion stationary

Time steady

Flow materials Gas / air

Air density 1.18415 kg/m^3

Dynamic viscosity 1.85508E-5 Pa-s

Flow type Couple

Equation of state Constant density

Viscous Regime Turbulence (Reynolds

averaged Navier-Stokes)

Reynolds averaged

turbulence

K-Epsilon turbulence

Ship speed 23knots

CFD simulation conditions for above water hull

analysis:

Mesh size: On container stacks and deck house- target

size 0.6m and minimum size 0.2m, on the deck and on the

above water hull- target size 0.8m and minimum size 0.2m.

The total boundary length is 1000m, and breadth also is 1000 m, the height is 245m

and the ship position at the centre of the

bottom surface. The length and the breadth

are the same because the ship is rotated

from 0 deg to 180deg. 5

Mesh/grid generation

Grid/mesh generation is the mostimportant task and valid mesh generation is

the most time consuming part in CFD

analysis.

The quality of the CFD analysis mostlydepends on the quality of generated mesh.

Mainly three types of mesh: structured,unstructured and hybrid. Here the

unstructured mesh and hybrid mesh are

used.

Generating the mesh type for CFDanalysis by Starccm+ is Polyhedral. In

analysis the volumetric control density is

2.5m.

The used numbers of prism layers are 4for 3 cm

6

Graphical presentation of CFD Simulation Result

The simulation results can be presented by

1. graphical from

2. tabular form

In graphical form the streamlines represent the air

flow and help to give us a

better understanding of the

numerical results

Gaps between container stacks can have a significant

influence on the resulting

forces

Graphical presentation of the

simulation result

7

Simulation Result as Tabular Form

The result of the pressure and shear forces on different stacks arepresented in the following table (ship speed 23 knots in head wind 20 knots)

The drag force acting on the different parts of the ship hull and container stacks

Part Pressure(N) Shear(N) Net(N)

------------------------------ ------------- ------------- -------------

DH -2.068550e+04 -1.023979e+02 -2.078790e+04

hull -1.369280e+04 -2.036224e+03 -1.572903e+04

Stack_1 -7.055177e+03 -5.296690e+01 -7.108144e+03

Stack_2 -4.209718e+02 1.244562e+01 -4.085262e+02

Stack_3 -2.034496e+04 -1.264827e+01 -2.035760e+04

Stack_4 1.518748e+04 -3.620133e+01 1.515128e+04

Stack_5 -1.888242e+04 -3.576994e+01 -1.891819e+04

Stack_6 1.559257e+04 -4.994998e+01 1.554262e+04

Stack_7 -2.733035e+04 -4.540582e+01 -2.737576e+04

Stack_8 2.933077e+03 -8.248552e+01 2.850592e+03

Stack_9 1.596127e+02 -8.651351e+01 7.309917e+01

Stack_10 -5.175854e+02 -8.462801e+01 -6.022134e+02

Stack_11 -7.120480e+03 -8.316351e+01 -7.203644e+03

Stack_12 -2.639828e+03 -7.487794e+01 -2.714706e+03

Stack_13 1.671533e+03 -4.370696e+00 1.667162e+03

Stack_14 9.267767e+02 2.337641e+00 9.291143e+02

------------------------------ ------------- ------------- -------------

Total: -8.221904e+04 -2.772820e+03 -8.499186e+04

Monitor value: -84991.85938N

Assign the container

stacks, deck house and

the hull

8

9Above water hull optimization

The forecastle deck is removed during the simulation in order to

investigate the stacks effect on the aerodynamic resistance properly.

The ship speed is 23 knots and wind speed is 20knots with

head wind condition.

The internal spaces among the stacks are 0.6 and 1.2m

The simulation results are taken from the M. Sc. project work done under Marine Technology, NTNU, Norway and partially financed by DNV

10

Comparison By applying:-

1.General form of stacks

2. By modifying 3 rear container stacks, for

considering accommodating the available spaces due

to remove the stacks

3.The 45o drag reduction surface with the front edge

of the first stack, with modifying rear stacks

4.Sloping upper surface including above modification

The simulation results are taken from the M. Sc. project work done under Marine Technology, NTNU, Norway and partially financed by DNV

Air resistance(KN)

1 103.6

2 96.84

3 85

4 69.86

Final drag force curve

0

50000

100000

150000

200000

250000

300000

350000

400000

0 20 40 60 80 100 120 140 160 180

Angle betweent the ship sailing direction and the wind direction

Dra

g f

orc

e

Full loaded conditionPartially loaded condition

Full loaded means all of the container stacks are present during simulation

Partially loaded means the container stacks 9 and 10 are removed during

simulation

For the angle between wind direction and the ship advance 140 and 30 the

drag forces are highest.

Streamlines & pressure of air on stacks

and on hull when incidence angle 0

Streamlines & pressure of air on stacks

and on hull when incidence angle 90

and the stacks 9 and 10 are removed

CFD simulation for different wind flow direction

11

Results of container stacks modification

For the reduction of the produced emission gases the counteraction may create other severe problems

This paper reviews the reduction in the production of the emission gases which is achievable by reducing the fuel consumption.

12

Due to optimization of the container stacks for 1000

nautical miles distance

Reduction of fuel consumption 2.83 tonnes

Reduction of emission gas CO2 about 6.6 tonnes

13

Conclusions

By applying the design optimization:-

The aerodynamic drag force can be reduced by attention to the layout and steamlineing of the container stacks

Due to increase in the spaces between containers the drag forces will also increase

The emission of exhaust gases produced from the fuel can be reduced by design optimization

The most important things are the proper knowledge and understanding about ship design optimization and that CFD simulation is

used properly. Interesting questions are:-

- Verification of the CFD results

- Size and resolution of the model

14

Summary of CFD results for this case study:-

A drag reduction surface at 45 on front row of containers reduced air flow resistance by 11.5%

By sloping the upper surface of the container stacks and avoiding large

gaps between stacks the air resistance could be reduced by about 15%

Streamlining of containers on the after deck behind the deck house

reduced the air resistance by about 6.5%

By design optimization a reduction of air resistance of about 33% was

achieved.

The air resistance was 3.2% of the total resistance for this design and

speed of ship leading to fuel and emissions reductions of ~ 1% .

15

Thank you for your attention !