Dynamic Response Prediction of Floating Offshore Wind

PO. 201 Turbine System using Fully Nonlinear Model

Dept. of Civil Engineering, The University of Tokyo
Takeshi Ishihara, Muhammad Bilal Waris, Ken Kagaya

Objectives Effect of heave plates and nonlinear mooring

The experimental model was tested without (D=8m)
Floating Offshore Wind Turbine System (FOWTS) is
and with heave plates (D=12m,16m).
made of light materials compared to conventional 4 4
offshore structures, which results in large motion under Surge [ H = 12m ] Heave [ H = 12m ]

Heave '2Z/H'
wind and wave load. Therefore, the nonlinearity of the 3 3

Surge '2X/H'
No Heave plate
12 m Heave plate
mooring stiffness and the dynamic interaction 2
16 m Heave plate
2
between wind turbine, floater and mooring system have
1
to be considered when dynamic response analysis of 1

FOTWS is performed. 0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Existing studies [1][2] on the dynamic motion of Catenary (left) and Tension Legged (right) mooring system Wave Period (sec) Wave Period (sec)
FOWTS are based on some simplifications, such as Response with different sized heave plates (no wind)
ignoring the dynamic interaction, assuming Total Lagrangian formulation with Newmark-  method
hydrostatic restoring force and linear mooring system, is used to solve the full elastic structure. Rayleigh • Heave plates reduced the heave response by up to
which may cause some errors. damping is considered with damping ratio of 0.5% more than 50% between 5-20 second wave period
The University of Tokyo has been developing a fully range. This is due to the shift of the resonance peak to
nonlinear FEM for the dynamic response prediction of Modeling of the FOWTS longer period caused by the increased added mass.
Component Description No. of Elements Type
FOWTS[3]. In this study, non-hydrostatic restoring force FOWTS with nonlinear catenary and tension legged
Wind Turbine Tower 13 Beam
model was applied and verified through comparison with mooring were modeled using the FEM and their response
Floater 109 Beam
hydrostatic model and water tank experiment. Then the to regular waves with no wind was simulated. The results
Experimental Elastic 4 Spring
effect of heave plates was investigated with this Setup were compared with linear mooring assuming restoring
Mooring Kevlar 24 Truss
updated model. Finally, based on this model, nonlinear force proportional to the displacement.
System Catenary Mooring 30 / line Truss
mooring system considering full dynamic coupling was Pre-stressed Beam 4 4
Tension Leg Mooring 10 / tether Surge [ Catenary ] Surge [ Tension Legged ]
modeled and their effects compared to linear mooring

Surge '2X/H'
3
system were simulated. Specification 3 Linear Linear
Nonlinear Nonlinear
Floater span 60m 2
2
Tower height 70m
Nonlinear FEM model 1
Sea depth 150m 1

Catenary chain length 660m

The equation of motion can be written as 0 0
0 10 20 30 40 50
Total pretension of TLP tendons 6870kN 0 10 20 30 40 50
Mx + Cx + Kx = f 4
Wave Period (sec)
2 10
-3
Wave Period (sec)
where Heave [ Catenary ] Heave [ Tension Legged ]
M T 0  CTT 0  K TT 0   xT 

Heave '2Z/H'
0 CTF K TF 3
-3

  Verification of the nonlinear model 1.5 10

M =  0 MF 0  , C = CFT CFF CFM  , K =  K FT K FF K FM  , x =  xF 
 0 M M   0 CMM   0 K MM  x  -3
0 CMF K MF  M 2 1 10
A water tank experiment was conducted to verify the
The subscripts refer to Turbine, Floater and Mooring FEM model. The scale was 1:100 following Froude 1 5 10
-4

system respectively. External force includes: similitude. Four horizontal mooring lines connected to 0
0 0 10
Gravitational force, Buoyancy force, Hydrodynamic elastic bands were used. Waves corresponding to 4m 0 10 20 30 40 50 0 10 20 30 40 50
Wave Period (sec) Wave Period (sec)
force, Restoring force, Seabed contact force and (rated state) and 12m (extreme state) were generated -3
4 2 10
Aerodynamic force. and the response of the floater was observed. Pitch [ Catenary ] Pitch [ Tension Legged ]
-3
3 1.5 10
• Hydrodynamic force on cylinder members is modeled
Pitch '2Lq/H'

using Morison’s equation modified by Sarpkaya & 4500 2 1 10

-3

Isaacson[4]. This force consists of added inertia force, 1500

Wave Height
-4
1 5 10
Froude-Krylov force and drag force. Meter 445

600 3000 0
fH = fHM + fHW + fHD Model 0 0 10
0
0 10 20 30 40 50 10 20 30 40 50
fHM = -M a x where M a  w  CM  1 A Tension Meter Elastic Band Wave Period (sec) Wave Period (sec)
fHW  wCM Au
1
Floater response for linear and non-linear mooring [H = 4m]
fHD   wCD Au - x u - x
2 • For th e c a te nary m oor i ng , t he l i n ear m od el
C D , CM are functions of Keulegan-Carpenter number. Experimental setup
overestimated the surge response. This is due to the
• Hydrodynamic force on column base consists of added 4 4
large dynamic component in the mooring force.
inertia force and damping force. Surge [ H = 12m ] Heave [ H = 12m ] • For the tension legged mooring, both models showed
Surge '2X/H'

3 Experiment
Heave '2Z/H'

3
fH = fHM + fHD Hydrostatic
Non-Hydrostatic
similar results. This is due to the strong initial tension
fHM = -M a x where M a  w  CM  1V 2 Dynamic Pressure Model
2 of the tendons resulting in relatively small nonlinearity.
fHD = CED x 1 1
Conclusions
where V is a half sphere volume of water proposed in 0
Haslum’s model[5]. Damping ratio is 15% following 0 5 10 15 20 25 30
0
0 5 10 15 20 25 30 • A fully coupled nonlinear FEM was developed to predict
Wave Period (sec) Wave Period (sec)
Srinivasan’s model[6] the dynamic response of FOWTS and it was
4 0.05 verified through water tank experiment.
• Restoring force: Pitch [ H = 12m ]
0.04
Mooring tension [ H = 12m ]
• Heave plates reduced the heave response by shifting
Pitch '2Bq /H'

DT/ p0 BS

3
Non-hydrostatic restoring force 0.03 the resonance peak to longer period.
fR NHM
= K R  x - η 2
• Simulation results showed small nonlinearity in tension
0.02
where  is the wave height. 1 l e g g e d m o o r i n g . H o w e v e r, t h e l i n e a r m o d e l
0.01

• Seabed contact force for nonlinear catenary mooring 0 0

overestimated the surge for catenary mooring due to
consists of frictional and normal force acting only on
0 5 10 15
Wave Period (sec)
20 25 30 0 5 10 15
Wave Period (sec)
20 25 30 the large dynamic component in the mooring force.
elements in contact with the sea bed. According to Ju References
et al [7] , it is expressed using penalty constant k, Response of the floater to regular waves (no wind)
frictional coefficient  and relative displacement in 1. J.M. Jonkman, Dynamic modeling and load analysis of an offshore floating wind turbine,
• Non-hydrostatic model agreed well with the Department of Aerospace Engineering Sciences, University of Colorado, Ph.D Thesis, 2007.
tangential and normal directions 2. A. Henderson, M. Patel, Rigid-Body Motion of a floating offshore wind farm, Int. Journal of
experiment for the surge, heave, pitch and the Ambient Energy, Vol.19, No.3, 1998, pp: 167-180.
 ft   2   U  3. P.V. Phuc, T. Ishihara, A study on the dynamic response of a semi-submersible floating
fc      k    mooring line tension. Hydrostatic model offshore wind turbine system Part 2: Numerical simulation, ICWE12, Australia 2007.
fn   1  V 
underestimates the heave response due to 4. T. Sarpkaya, M. and Isaacson, Mechanics of wave forces on offshore structures, Van
Norstrand Reinhold,1981.
• Aerody namic for c e i s modeled us ing Q uas i - underestimation of vertical restoring force. 5. H.A. Haslum, Alternative Shape of Spar Platforms for Use in Hostile Areas, Offshore
Technology Conference, 1999.
steady theory, Blade element theory and Momentum • These results verify the hydrodynamic force and 6. Srinivasan, N., Chakrabarti S. and Radha R., Damping controlled response of a truss
pontoon semisubmersible with heave plates, Proceedings of the 24th International
theory considering blade tip loss, hub loss and tower response evaluation in the developed model. Conference of OMAE, Halkidiki, Greece, 2005.
shadow. However, wind is not considered in this study. Similar results were seen for the 4m case as well. 7. Ju S.H., Stone J.J. and Rowlands R.E., A new symmetric contact element stiffness matrix for
frictional contact problems, Computers and Structures, Vol. 54, No. 2, 1995, pp: 289 – 301.

EWEA OFFSHORE 2011, 29 November – 1 December 2011, Amsterdam, The Netherlands