15 10 5 0 5 10 15

15

10

5

0

5

10

15

Leg distance x [m]

Lattice tower 120m, 4 legs, 10 sections

Leg

dist

ance

y [m

]

Wind Power R&D seminar Deep sea offshore wind power, 20.-21. January 2011, Trondheim, Norway

LOADS AND DYNAMICS IN LATTICETOWER SUPPORT STRUCTURESFOR OFFSHORE WIND TURBINES

PhD candidate: Daniel ZwickSupervisor: Geir Moe Department of Civil and Transport Engineering

BACKGROUND

The extremely ambitious political goals concerning extensive use of offshore wind energy result in an in-tense demand of research and development in this field. As an example, round 3 in UK could mean a need to install several thousands of offshore wind turbines within the next ten years. To be able to fulfil this goal, components for offshore wind farms has to be produced by mass production techniques and within reasonably short fabrication time.New node concepts might be of interest for more au-tomated production of lattice towers. As a basis for such an investigation, loading and dynamic re-sponse by focusing on design of the nodes has been analysed with HAWC2 in this study.

SUPPORT STRUCTURE CONCEPTS

Where offshore wind turbines are planned to be in-stalled in the intermediate water depths of 30-70m, bottom-fixed support structures might be used. One promising concept is the lattice tower type, due to less material use compared to other concepts like monopile or tripod structures. A lattice topology could be used for the entire support structure be-tween sea bottom and turbine nacelle or for the lower part of the tower only.

Bottom-fixed support structure concepts forthe intermediate water depth of 30-40m

LATTICE TOWERS

Lattice towers are assembled from steel tubes, where legs and bracings are welded together in tubular joints. Legs and bracings are connected in K-joints, while bracings in the planes between the legs are connected in X-joints.

Joint geometry of nodes in lattice towers

NODE ANALYSIS WITH HAWC2

A lattice tower support structure with 84 beam ele-ments was modelled and analysed with HAWC2. Wind turbine and rotor configuration were taken from the NREL 5MW baseline turbine.

NODE ELEMENT FORCES

NODE ELEMENT MOMENTS

FATIGUE ANALYSIS

MEMBER DIMENSIONS

The initial tower design of this study was analysed with constant leg and bracing dimensions over the tower height. As expected, results from the fatigue analysis show that dimensions for the legs has to be increased towards sea bottom, while bracing dimen-sions hast to be increased towards tower top. First calculations were based on a traditional node design with circular members intersecting each other. The shown load results will be used for the futher analy-sis of new node designs, suitable for mass produc-tion of lattice towers.

Results from HAWC2 are obtained in time domain. The figure to the left shows an analysis of a com-plete K-joint in one leg at a specific node. The distri-bution of mean forces in one leg over the tower height is shown to the right, with standard deviation and min/max range. Absolute forcesin z-direction are decreasing towards the tower top.

Mean forces in the bracing X-joints are more orless stable over the tower height, but standard deviation and min/max range are increas-ing towards the tower top.

From the same analysis, results for all member mo-ments were extracted. Mean values over the tower height were found to be close to zero in the legs.However, the range of min/max values is increasing strongly towards the tower top. Moments in the high-est tower nodes are dependent onthe connection design of tower and nacelle.

For the bracing members in X-joints, only smallmoments were found, varying around a zero mean. Bracing members are mainly loaded by axial forces.

OBJECTIVES

New node concepts for lattice towers will be devel-oped for the following purposes:

- lower total production costs - faster production, towards mass production - more automated production - more reliable welding results - prefabrication of components

If the complex fabrication of lattice towers can be solved in an effective way, this type might be a pre-ferred solution for support structures in the future.

15 10 5 0 5 10 150

20

40

60

80

100

120

Leg distance x [m]

Lattice tower 120m, 4 legs, 10 sections

Tow

er h

eigh

t z [m

]

0 100 200 300 400 50010

5

0

5

10X legs 0sti, c1, d2, 1120

Fx c

oo: l

ocal

[kN

]

0 100 200 300 400 50010

5

0

5X bracings 0sti, c1, d2, 1120

0 100 200 300 400 50015

10

5

0

5

10Y legs 0sti, c1, d2, 1120

Fy c

oo: l

ocal

[kN

]

0 100 200 300 400 5004

2

0

2

4

6Y bracings 0sti, c1, d2, 1120

0 100 200 300 400 5006000

4000

2000

0

2000Z legs 0sti, c1, d2, 1120

Fz c

oo: l

ocal

[kN

]

Time [s]

leg1 5 1

leg1 4 2

Node forces time series in Kjoints, node 5

0 100 200 300 400 500400

200

0

200

400Z bracings 0sti, c1, d2, 1120

Time [s]

bra51 1 1bra52 1 1bra44 2 2bra47 2 2

8000 6000 4000 2000 0 20000

50

100

Z leg1 0sti, c1, d2, 1120

Tow

er h

eigh

t [m

]

Force [kN]

500 0 5000

50

100

Z braX1 in 12plane 0sti, c1, d2, 1120

Tow

er h

eigh

t [m

]

Force [kN]

100 0 1000

50

100

z leg1 0sti, c1, d2, 1120

Tow

er h

eigh

t [m

]

Moment [kNm]

4 2 0 2 40

50

100

z braX1 in 12plane 0sti, c1, d2, 1120

Tow

er h

eigh

t [m

]

Moment [kNm]

0 100 200 300 400 50050

0

50

100X legs, node 5 0sti, c1, d2, 1120

Mx

coo:

loca

l [kN

m]

0 100 200 300 400 50020

10

0

10

20X bracings, node 5 0sti, c1, d2, 1120

0 100 200 300 400 50040

20

0

20

40

60Y legs, node 5 0sti, c1, d2, 1120

My

coo:

loca

l [kN

m]

0 100 200 300 400 50015

10

5

0

5

10Y bracings, node 5 0sti, c1, d2, 1120

0 100 200 300 400 50050

0

50

100Z legs, node 5 0sti, c1, d2, 1120

Mz

coo:

loca

l [kN

m]

Time [s]

leg1 5 1

leg1 4 2

Node moments time series in Kjoints, node 5

0 100 200 300 400 5004

2

0

2

4Z bracings, node 5 0sti, c1, d2, 1120

Time [s]

bra51 1 1bra52 1 1bra44 2 2bra47 2 2

initial tower design HAWC2 analysis results in time domain

0 50 100 150 200 250 300 350 400 450 500400

200

0

200

400Z bracings 0sti, c1, d2, 1120

Fz c

oo: l

ocal

[kN

]

Time [s]

bra51 1 2bra54 1 2bra51 2 1bra54 2 1

rainflow counting Palmgren-Miner, S-N curve lifetime of the joint

424 426 428 430 432 434 436 438 440

300

200

100

0

100

200

300

400

Rainflow series of 185Fzbra10722

Time [s]

Forc

e [k

N]

0 5000 100000

50

100

150Z leg1 5 1 0sti, c1, d2, 1120

num

ber o

f cyc

les

FZ

amplitude [kN]

104

105

106

107

108

109

101

102

103

SN curve for tubular joints in seawater

Number of cycles

Stre

ss r

ange

[MPa

]

101

102

103

0

50

100

Z legs 0sti, c1, d2, 1120

Tow

er h

eigh

t [m

]

lifetime [years]

leg1 X 1

leg1 X 2

20 years

101

102

0

50

100

Z bracings 0sti, c1, d2, 1120

Tow

er h

eigh

t [m

]

lifetime [years]

braX1 1 1braX2 1 1braX4 2 2braX7 2 220 years

1

ki

i i

nDN