Master thesis: Development of a practical tool to determine the hull damping of modern ship hull forms
Bilel SAAD
La Spezia 17.02.2014
Motivation
2 EMSHIP UNIVERSITÄT ROSTOCK
Roll damping prediction:
Why ?
Response in rough weather
Design of Stabilisation equipments
Aim ?
Modern hull forms
New methodology for
roll damping prediction
Alternative to Experimental model tesing
17.02.2014
Obstacles and Challenges
EMSHIP UNIVERSITÄT ROSTOCK 3
High Non linearity Equivalent linear roll
damping coefficient
Obstacles Remedy
Energy dissipation
Spring mass system Equation
17.02.2014
Scenario of the Numerical simulations
4 Simulation tools
CFD:
Potential flow theory: PDstrip
Component Analysis Method: IKEDA Original Method
Component Analysis Method: IKEDA Simple
formulation Method
Empirical method:
Miller method
3 Test
case ships
16000-18000 TEU Ultra large container ship
(Container ship N°1)
8000-9000 TEU Large container ship
(Container ship N°2)
DTMB 5415: ARLEIGH BURKE-class
destroyer
(US Navy Ship )
4 EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Potential flow theory simulation: Pdstrip solver
• Potential flow solver : Strip theory+ Panel method
• Wave damping accurately predicted
• Forces on fins and appendages
• Output => Transfer function(RAO)
PDstrip => RAO 2DRoll=> Regression analysis
Roll damping coefficient
B44
5
PDstrip Numerical Simulations For the US Navy ship
Estimation of roll damping from the transfer function (RAO)
Beam seas ,0 forward speed & T=6.15 m
6
0
2
4
6
8
10
12
14
0 0,5 1 1,5 2 2,5 3 3,5 4
Ro
ll R
AO
(η
44
/kA
)
λ/L
PDSTRIP DTMB 5415 Stationary test (0 Fw speed) Beam seas
Experimental RAO DTMB 5415 Stationary test (0 FW Speed, Beam seas)
Round bottom assumption => Lower eddy making damping
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Shape of the bottom of the US Navy ship model DTMB 5415
7
Sharp corners
and high rake
zones
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Sensitivity analysis for PDstrip RAO regarding the GM value for the container ship N°1
8
Roll transfer function(RAO) in (°/m) versus wave frequency in (rad/s)
stern quartering (45° from the stern) , 21 kts, GM=9m
0
0,5
1
1,5
2
2,5
3
0 0,2 0,4 0,6 0,8 1
Ro
ll R
AO
in (
de
g/m
)
Wave frequency in (Rad/s)
Experimental PDstrip RAO, 45°, GM=9m , 21 Kts PDStrip, 45°, GM=9.1m,21 kts
Difference due to 10 cm change in GM
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Component analysis method: IKEDA original and IKEDA simple methods
9
Component analysis method
Discretisation of the damping coefficient
Be = Bw + BF + BBK + BE + BL
IKEDA Original method
- Input:
Hull Form( Offsets), Bw (wave)
- Output:
BL (Lift), BE (Eddy), BF (Friction), BBK (Bilge Keel)
IKEDA simple method
- Input:
Hull Form (B,T, Lpp…), BL (speed)
- Output:
Bw (wave), BE (Eddy), BF (Friction), BBK (Bilge Keel)
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Component analysis method for Ultra large Container ship N° 1 :IKEDA Original and Simple Methods
10
90° heading (Beam seas), 21 Kts forward speed, GM=2.9 m
0
0,001
0,002
0,003
0,004
0,005
0,006
0,007
0,008
0,009
0,01
0 2 4 6 8 10 12
No
nd
imen
sio
nal
Dam
pin
g
Roll amplitude in (°)
Ikeda original method Roll decay IKEDA Simple method
Slope of damping= f(Roll amplitude) => Non linearity
Difference : BK drag ( HIGH period)
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Component analysis method for the US Navy ship: IKEDA original method
Comparison of Ikeda original method results for DTMB 5415 with roll decay test results for two
different initial inclination angles( 20 ° and 15 °) at Fn=0.28.
11
0,0065
0,007
0,0075
0,008
0,0085
0,009
0,0095
0,01
0,0105
0 5 10 15 20
No
nd
ime
nsi
on
al d
amp
ing
coef
fici
en
t
Roll amplitude (°)
Roll decay fi0=15°, Fr=0,28 Roll decay fi0=20°, Fr=0,28 Ikeda original method Fn=0.28
Initial Inclination =20 °
Initial Inclination =15 °
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Miller method
• Regression analysis => US Navy ships
• Early design stage
12
Miller method
Step1: Damping at zero forward speed
Non linear formula
Step 2 : Forward speed effect
Input of Froude number
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Miller Method Simulations For (ultra large) Container Ship N°1
Forward speed 21Kts & Draft T=10.5m
13
0,000
0,001
0,002
0,003
0,004
0,005
0,006
0,007
0,008
0,009
0 2 4 6 8 10 12 14
No
nd
ime
nsi
on
al d
amp
ing
Roll amplitude in deg
Ikeda method,21kts,90° Roll decay Miller Method at 21Kts & T=10.5m
Good agreement for smaller angles
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Miller method simulation for the US Navy ship
Forward speed 20 Kts & Draft T= 9.97m
14
0
0,002
0,004
0,006
0,008
0,01
0,012
0 2 4 6 8 10 12 14 16 18 20
No
nd
imen
sio
nal
dam
pin
g co
effi
cien
t
Roll amplitude (°)
Roll decay fi0=15°, Fr=0,28 Roll decay fi0=20°, Fr=0,28 Miller Method 20Kts & T=9.97
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Conclusion
EMSHIP UNIVERSITÄT ROSTOCK 15 17.02.2014
Established methodology for roll damping prediction
16
Step 1: Selecting the prediction method according to ship type:
•Potential theory: Pdstrip => - Ships with high Bw
•Component analysis-=>Low VCG (No car carrier , No large passenger vessel)
•Miller method : => Slender ships: Navy ships (Low CB )
Step 2: Approximating the damping coefficient with the fast Miller method
•Order of magnitude
• Fast method
Step 3: Running the selected method
•Benefit from the complementarity of certain methods
No tool available for all kinds for ships
EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014
Complementarity and different possible combinations
Combination 1
Pdstrip
Wave component Bw
IKEDA original method
BF(Friction) + BL(Lift) + BBK(Bilge keel)
Combination 2
Pdstrip
Roll response
2Droll:
Damping component from regression
analysis
Combination 3
IKEDA original method
Lift component BL
IKEDA simple method
BF(Friction)+BBK(Bilge keel) + BW(wave)
Combination 4
IKEDA simple method
Wave component Bw
IKEDA original method
BF(Friction)+BL(Lift)+ BBK(Bilge keel)
17 EMSHIP UNIVERSITÄT ROSTOCK 17.02.2014