The effect of a speed exclusion zone and active tower dampers on an upwind fixed-hub two-bladed 20 MW wind turbine

F. Anstock1, M. Schütt1 and V. Schorbach1

1Hamburg University of Applied Sciences Deep Wind Conference, 13.01.2021

Who are we?

2

Between

and

Funded by

Cooperation project:

“X-Rotor – two-bladed wind turbines”20 MW turbines of the next generation

Source: Levin Schilling

Pro Contra

• Cheaper rotor and drivetrain

• Faster and easier erection

• Less components to maintain

• Better access by helicopter

• Lower turbine head mass

• More noise• More unpleasant looks• Lower power coefficient (Cp)

• More harmful dynamics

Pro Contra

• Cheaper rotor and drivetrain

• Faster and easier erection

• Less components to maintain

• Better access by helicopter

• Lower turbine head mass

• More noise• More unpleasant looks• Lower power coefficient (Cp)

Extend rotor size by 2%1

• More harmful dynamics Today better controllable

(active or passive) Size effect

Offshore:

Introduction – Why two-bladed turbines?

31 F. Anstock et. al., A new approach for comparability of two- and three-bladed 20 MW offshore wind turbines,

Journal of Physics: Conference Series, 2019

Three- and two-bladed reference turbines – design process

4

3-Bladed INNWIND 20 MW as reference turbine

Redesign the blades with similar aerodynamics1 and ~50% higher strength and buckling resistance2

2-Bladed reference turbinewith equal power curve

1 F. Anstock et. al., A new approach for comparability of two- and three-bladed 20 MW offshore wind turbines, J. Phys.: Conf. Ser., 20192 M. Schütt et. al., Progressive structural scaling of a 20 MW two-bladed offshore wind turbine rotor blade examined

by finite element analyses, J. Phys.: Conf. Ser., 2020

5

Three- and two-bladed references – Objective PI-gain tuning1

0.01

0.05

0.1

0.2

0.3

0.6

11.5

2

0.94

0.96

0.98

1

1.02

1.04

0.20.6

11.4

1.82.2

2.6

I-ga

in f

acto

r

J_C

CC

P-gain factor

CCC for 3B ref with NTM at 15 m/s and 6 seeds mean

0.94-0.96 0.96-0.98 0.98-1 1-1.02 1.02-1.04 1.04-1.05

Feasible minimum forΩ𝑚𝑎𝑥 = Ω𝑟𝑎𝑡𝑒𝑑 ∙ 1.08

1 F. Anstock et. al., A control cost criterion for controller tuning of two- and three-bladed 20MW offshore wind turbines, J. Phys.: Conf. Ser., 2020

The control cost criterion (CCC) estimates the impact on costs of pitch activity, energy yield, tower, hub and blade loads (for DLC 1.2) to achieve a good compromise between conflicting objectives

6

Three- and two-bladed references – first load comparison NTM 15 m/s

The described procedure leads to following results for 15 m/s NTM and a rated tip speed of 90 and 100 m/s for the 2-bladed turbines1:

Lower loads for higher tip speed

-50%

50%

150%

250%

5 7 9 11 13 15 17 19 21 23 25

rel.

dif

fere

nce

wind speed in m/s

(2B101/3B)opt -1 for tower bending loads of DLC 1.2

1 F. Anstock et. al., A control cost criterion for controller tuning of two- and three-bladed 20MW offshore wind turbines, J. Phys.: Conf. Ser., 2020

-50%

0%

50%

100%

150%

200%

250%

300%

350%

400%

5 7 9 11 13 15 17 19 21 23 25

rel.

dif

fere

nce

wind speed in m/s

2B101 ref 2B101 with speed excusion zone 2B101 with speed exclusion zone and dampers

7

Speed exclusion zone – challenging dynamics

Step 1: Speed exclusion zone

relative differences of DLC 1.2 tower loads compared to 3B with speed exclusion zone and dampers

1P

2P

3P

8

Speed exclusion zone – Campbell diagram

0.000

0.050

0.100

0.150

0.200

0.250

0.300

0.350

0.400

4 5 6 7 8 9 10 11 12 13 14

freq

uen

cies

in H

z

wind speed in m/s

1P -- 3B ref

2P -- 3B ref

3P -- 3B ref

1P -- 2B101 ref

2P -- 2B101 ref

3P -- 2B101 ref

1st StS tower eigenfrequency

1st FA tower eigenfrequency

3P of 3-Bladed turbine

2P of 2-Bladed turbine

Tower frequency excitation by passing blades shifts:

Blade-tower-interference for 2-Bladed turbine at higher wind speeds with more power and higher occurrence

9

Speed exclusion zone – how does it work? Basic procedure

Use 𝜔avoid ± safety gap as reference variable for PI-torque controller

rotation speed 𝜔

Is 𝜔 close

to𝜔avoid?

Either direct exit

or cross𝜔avoid rapidly by spline-function

Controller continues

YesNo

No

Yes

𝜔avoid ෝ=𝑡𝑜𝑤𝑒𝑟 𝑒𝑖𝑔𝑒𝑛𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦

𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑏𝑙𝑎𝑑𝑒𝑠60

Generatortorque inside

min/maxregion?

70

90

110

130

150

170

0 100 200 300 400 500 600Ge

ne

rato

r To

rqu

e in

kN

m

time in s

3B with and without speed exclusion zone with NTM at 5 m/s

3B ref 3B with speed exclusion zone

22.2

2.4

2.62.8

3

3.23.4

3.6

3.8

4

0 100 200 300 400 500 600

rota

tio

n s

pe

ed

in r

pm

time in s

3B ref 3B with speed exclusion zone

-120

-80

-40

0

40

80

120

0 100 200 300 400 500 600

tow

er

bas

e S

tSm

om

ent

in M

Nm

time in s

3B ref

3B with speed exclusion zone

10

Speed exclusion zone – how does it work? Example

𝜔 safety zone𝜔avoid (3.3 rpm)

Threshold

Wave-like spline function

Back to normal mode

Huge difference in tower fatigue

70

90

110

130

150

170

0 100 200 300 400 500 600Ge

ne

rato

r To

rqu

e in

kN

m

time in s

3B with and without speed exclusion zone with NTM at 5 m/s

3B ref 3B with speed exclusion zone

22.2

2.4

2.62.8

3

3.23.4

3.6

3.8

4

0 100 200 300 400 500 600

rota

tio

n s

pe

ed

in r

pm

time in s

3B ref 3B with speed exclusion zone

-120

-80

-40

0

40

80

120

0 100 200 300 400 500 600

tow

er

bas

e S

tSm

om

ent

in M

Nm

time in s

3B ref

3B with speed exclusion zone

10

Speed exclusion zone – how does it work? Example

𝜔 safety zone𝜔avoid (3.3 rpm)

Threshold

Wave-like spline function

Back to normal mode

Huge difference in tower fatigue

3B: Tower DEL1 reduced by -9.9%and energy yield1 by -0.03%

2B101: Tower DEL1 reduced by -33.8%and energy yield1 by -0.17%

1 DLC 1.2 with 0°, +-8° yaw and Rayleigh distribution with 11.4 m/s mean wind speed

-50%

0%

50%

100%

150%

200%

250%

300%

350%

400%

5 7 9 11 13 15 17 19 21 23 25

rel.

dif

fere

nce

wind speed in m/s

2B101 ref 2B101 with speed excusion zone 2B101 with speed exclusion zone and dampers

12

Active tower dampers – challenging dynamics

relative differences of DLC 1.2 tower loads compared to 3B with speed exclusion zone and dampers

Step 1: Speed exclusion zone

Step 2: Active FA and StS dampers

13

Basic active fore-aft (FA) tower damper

Measure fore-aft acceleration andintegrate to FA velocity

Apply tower FA velocity signal with a gaindirectly on demanded pitch signal

-0.2

0

0.2

0.4

5

7

9

11

0 20 40 60 80

nac

elle

FA

vel

oci

ty in

m/s

pit

ch in

deg

time in s

pitch ref pitch with FA damping nacelle FA velocity

14

Basic active fore-aft (FA) tower damper

Reasonable gain for 3B is 0.03:

Tower DEL reduced by -4.57%,energy yield by -0.0001% andBlade My (flapwise) DEL increased by 0.46%

Reasonable gain for 2B101 is 0.125:

Tower DEL reduced by -26.57%,energy yield by -0.08% andBlade My (flapwise) DEL increased by 10.9%

Almost no energy loss present

Less benefits for the three-bladed turbine’s tower, but blade loads almost unchanged

-40%

-30%

-20%

-10%

0%

10%

20%

0 0.05 0.1 0.15 0.2

rel.

to u

nd

amp

ed

FA damping gain

2B101 FA damping for 11-25 m/s

-40%

-30%

-20%

-10%

0%

10%

20%

0 0.05 0.1 0.15 0.2

rel.

to u

nd

amp

ed

FA damping gain

3B FA damping for 11-25 m/s

tower My' DEL

blade My DEL (flapwise)

energy yield

15

Basic active side-to-side (StS) tower damper

Measure nacelle roll acceleration and integrate to the roll velocity

Apply nacelle roll velocity with a gaindirectly on demanded generator torque

-0.1

0

0.1

0.2

0.3

26000

26500

27000

27500

0 20 40 60 80

nac

elle

ro

ll ve

loci

ty in

deg

/s

gen

erat

or

torq

ue

in k

Nm

time in s

generator torque ref generator torque with StS damping nacelle roll velocity

16

Basic active side-to-side (StS) tower damper

The StS damping by generator works well

Has only a small effect on the combined tower DEL My‘ for 2B101 and 3B in full load operation

During partial load the influence is bigger.

(Reasonable) gain for 3B is 0.05:

Tower Mx DEL reduced by -7.7%,and combined tower DEL by -0.21%;energy yield increases by +0.01%

-30%

-25%

-20%

-15%

-10%

-5%

0%

5%

0 0.1 0.2 0.3 0.4

rel.

to u

nd

amp

ed

FA damping gain

3B StS damping for 11-25 m/s

tower Mx DEL (StS)

blade My DEL (flapwise)

tower My' DEL

energy yield

-30%

-25%

-20%

-15%

-10%

-5%

0%

5%

0 0.1 0.2 0.3 0.4

rel.

to u

nd

amp

ed

FA damping gain

2B101 StS damping for 11-25 m/s

(Reasonable) gain for 2B101 is 0.1:

Tower Mx DEL reduced by -14.9%and combined tower DEL by -0.47%;energy yield increases by +0.02%

< 0.5 %

0.5

1.0

1.5

2.0

2.5

3.0

Tower My'

3B ref 3B with dampers or exclusion zone 2B101 ref 2B101 with dampers or exclusion zone

17

Final load comparison – fatigue by DLC 1.2

0.50

0.75

1.00

1.25

1.50

Blade My

rel.

to 3

B

0.90

0.95

1.00

1.05

1.10

Energy yield

Huge differencedue to dampersand speedexclusion zone

34% from exclusion zone

27% from FA damper

9% from StS damper

18

Summary and Conclusions

The proposed control features reduce the two-bladed turbine‘s tower loads already from 170% to 28% above an equivalent three-bladed turbine. With teetering, MPC, IPC or free-yaw, this difference might shrink even more, potentially rectifying the largest drawback of two-bladed turbines – the more harmful dynamics.

Simple collective pitch driven tower fore-aft damping and side-to-side damping by the generator torque reduce tower fatigue efficiently by another 36% for the two-blade turbine.

A speed exclusion zone can be a good countermeasure against 2P or 3P excitation of the tower eigenfrequency, reducing tower fatigue (loads) by 10% for three-bladed and 33% for two-bladed.

Thank you for yourattention!

Fabian Anstock, M.Sc.

Research Associate

Project: X-Rotor – two-bladed wind turbines

T +49 40 428 75 8768fabian.anstock@haw-hamburg.de

HAMBURG UNIVERSITY OF APPLIED SCIENCESCompetence Center for Renewable Energy

and Energy EfficiencyBerliner Tor 21 / 20099 Hamburg

haw-hamburg.de

19

Source: Levin Schilling

Funded by

&