Offshore Wind Farm Design: Operation and Maintenance

1

Module 11: Operation & Maintenance

Gerard van Bussel

section Wind Energy

OPERATION & MAINTENANCE


Typical contribution to energy cost

Decommissioning3%

O & M

23%Investment

74%

Opti-OWECS study (1998)

Investment &

~ 75 %

Decommissioning

25 - 30%O & M

CA-OWEE (2001)

2


maintenance

correctivemaintenance

preventivemaintenance

periodicpreventive

maintenance

conditionbased

maintenance

conditionjudging

conditionmonitoring

batch wisecorrective

maintenance

correctivemaintenanceon demand

calendarbased

opportunitybased

repair service

Maintenance Concepts


Tacke TW600 Enercon 40 Vestas V39/500

number of turbines 25 26 59

events/year events/year events/yearLightning 0 0.03 0Blade 0.76 0.42 0.32Rotor Brake 0 0 0Pitch Mechanism 0 0.30 0.03Brake 0.08 0 0Shaft/Bearing 0.04 0.03 0Gearbox 0.16 0 0.03Generator 0 0.03 0.33Hydraulic 0.32 0 0.27Yaw System 0.32 0.23 0.08Anemometry 0 0 0.01Electronics 0.04 0.42 0.33Electric 0.20 0.69 0.30Inverter 0 0 0Sensors 0.08 0.07 0.18Other 0.20 0.38 0.37Overall Total 2.2 2.6 2.25

Failure frequencies 500kW class

Failure frequencies 2.2 2.6 2.25For larger machines (onshore) => 2.2 /year

3


Component Failure frequency(failures/year)

Shaft & Bearings 0.02Brake 0.05Generator 0.05Parking Brake 0.05Electric 0.14Blade 0.16Yaw System 0.23Blade tips 0.28Pitch Mechanism 0.28Gearbox 0.30Inverter 0.32Control 0.34Total 2.20

Failure frequencies multi MWW class

Total of all components: 2.20 failures/year


reliability(failures/year)

maintainability(MTTR)

serviceability(PM demand)

accessibilityof the site

maintenancestrategy

theoreticalavailability

actualavailability

actualavailability

Reliability, Availability, Maintainability, Serviceability

4


A measure for availability

time

staterunning

failedMTTF

Mean Time To FailureMTTR

Mean Time To Repair

Availability =MTTF

MTTR + MTTF


Experienced Availability

1996 1997 199890.00%

92.00%

94.00%

96.00%

98.00%

100.00%

Tuno Knob availability

1996 1997 199890.00%

92.00%

94.00%

96.00%

98.00%

100.00%

Tuno Knob availabilityTuno Knob (inshore,Denmark)

97%

5


Present maintenance demand

Visits perturbineperyear

Unplanned

1

3

0Planned

2

4


Oct Nov Dec Jan Feb Mar Apr May June July TotalNominalWorking Days

23 21 20 22 20 22 19 19 21 23 210

Bad Weather(Days)

1 1 0 2 4 5 4 0 0 0 17

Bad Weather(o.5 Day)

5 2 2 4 1 3 1 0 0 0 18

Lack ofTransport

0 2 0 0 1.5 4 6 1 0 0 14.5

InaccessibleDays

3.5 4 1 4 6 10.5 10.5 1 0 0 40.5

% Total Time 15.2 19 5 18.2 30 47.7 55.3 5.3 0 0 19.3

Accessibility of site (Vessel)

Vindeby DK inshore

6


Horns Rev versus Vindeby Accessibility

0.00%

20.00%

40.00%

60.00%

80.00%

100.00%

120.00%

jan

feb

mar ap

r

may jun jul

aug

sep

oct

nov

dec

Frac

tion

of ti

me

6 hours12 hours24 hoursVindeby

Accessibility of site (Vessel)Vindeby(inshore,Denmark)

Horns Rev(near shore,North Sea)


Means of crew transport

• Helicopter- fast- expensive - helipad- large operational

window

• Tender vessel- fairly slow- cheap- boat landing - medium

window

7


A “gol” boatHarbour pilots

A tender vesselFor crew transport

Means of crew transport 2


With a Zodiacfor landing

A tender vesselFor crew transport


8


With a Zodiacfor landing ??With a Zodiacfor landing ??




Crew transportby helicopter

9

Module 11: Operation & MaintenanceTrends: Access methods

Catamaran landing vessel

SWATH@A&R / Abeking &Rasmussen

Module 11: Operation & MaintenanceTrends: Access methods

Flexible gangway:

OAS: P&R systems / Reinout Prins

10


Trends: Access methods

OAS: P&R systems / Reinout Prins

Flexible gangwayFlexible gangway


Trends: Access methods

AmpelmannAmpelmann

11


Cost comparison transport Vessel/Helicopter

2 MW Wind turbine farm

010002000300040005000600070008000

0 10 20 30 40 50 60 70 80

Distance Mantenance base to offshore wind farm [km]

Cos

ts [e

uro]

VesselHelicopterVessel+DowntimeHeli+Downtime


Maintaining Horns Rev:– Access by boat: Winter 02/03: 5/7 days

» Winter 03/04: 1/7 days– Helicopter: 6/7 days

– Vestas responsible for crew (60 people)Elsam for transport (6 people)

– 75.000 transfers in 1.5 years (2 x /day/turbine)

Experiences in the real world

12


Experiences in the real world

Maintaining Horns Rev:

Reasons:– Design not well adapted for offshore– Strategy not optimal– Onshore crew– Sophisticated alarms, but what does it mean?


• Jack-up barge• Crane vessels• Helicopter

• Jack up vesselassisted with built-in facility (in wind turbine)

Lifting equipment

13


Installation & O&M lifting facilities

Utgrunden Wind farm, Sweden A2Sea Ocean Hanne at Horns Rev Denmark

Jack up vessels

Module 11: Operation & MaintenanceInstallation and O&M lifting facilities

A2Sea Ocean Hanne at Horns Rev and at NystedDenmark

Jack up vessels

14


Internal cranes

Picture: Enron (Utgrunden, Sweden)

Picture: Nordex N80 Offshore

Hoisting outside

Trends: Lifting at wind turbine


Trends: Installation

Ballast-Nedam NEG-Micon Dowec project

15


Maintenance strategies

• PM and CM on demand(onshore practice)

• Opportunity based maintenance(PM when CM is demanded)

• Condition based maintenance(PM and CM only when demanded)

• No maintenance/ batch maintenance


• PM and CM on demand(reduced PM demand, increased reliability)

• Opportunity based maintenance(flexible PM interval, increased reliability)

• Condition based maintenance(extensive condition monitoring)

• No maintenance/ batch maintenance(only feasible when failure freq. < 0.2 /year)

Maintenance strategies

16


50%

60%

70%

80%

90%

100%

40 % (remoteoffshore)

60 %(offshore)

80 %(nearshore)

100 %(onshore)

highlyimproved

improved

onshoredesign

Accessibility

Ava

ilabi

lity ReliabilityReliability

Importance of Reliability and Accessibility


Importance of (improved) Reliability

50%

60%

70%

80%

90%

100%


60 %(offshore)

80 %(near shore)

100 %(onshore)

OffshoreadaptedA

vaila

bilit

y

Vessel accessibility

50%

60%

70%

80%

90%

100%


60 %(offshore)

80 %(near shore)

100 %(onshore)

Offshoreadapted

Vessel accessibility

Horns Rev(North Sea)

Tuno & Vindeby(DK inshore)

Offshoredesigned

17


Assessing reliability, availability and O&M in the design process

Systems

specification

Conceptual design

Feasibilitystudy

RAMS targets on system

level

RAMS targets

sub-system level

RAMS targets on component level

RAMS targetssub-component level

Design evaluation

Final design

Expert system approach

Waiting time analysis

Full FMECA analysis

Monte Carlo Simulations


Probabilistic Waiting time analysisTime To Repair

IJmuiden Munitie Stortplaats

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 100 200 300 400 500 600 700 800 900 1000

time [hrs]

CDF T_mission: 24 h

T_mission: 72 h

T_mission: 168 h

Time [hrs]

Chance[%]

ECN/ Luc Rademakers

18


50 100 1500

20406080

100120

Max.availability

99.0 %98.0 %

97.0 %

Symbols: simulations

0No access percentage

Ava

ilabi

lity

[%]

Availability as Function of Maximum(onshore) Availability and Storm Percentage

Trend lines in expert system


• Analysis of complex stochastic processes• Failure simulation of wind turbines• Storm simulation for OWECS accessibility

• Availability estimates for OWECS

• O&M costs estimates for OWECS

Monte Carlo simulations

19


• systematic brake down of:- functional components- hardware component

• analysis of:- effects of all kinds of failures upon

functioning- criticality of failure

(how does failure affect costs/environment)

FMECA: Failure Mode Effect and Criticality Analysis


Reliability vs. turbine design

• Turbine design gets more complex:– Three bladed, variable speed pitch

control– Doubly fed generators, Inverters

BUT • Offshore environment demands a

robust, lean design:– Two blades !?– Stall control !??– Low speed or Direct drive generator !?

20


Recent wind turbine failures

NEG Micon

YttreStengrunden

V 80 at Tjaereborg

Nordex N 80Blyth offshore


Context of the project

• Consortium of industries and institutes• Project: ~ 500 MW offshore wind farm• Location: North Sea > 12 mile zone • Concepts of 5 MW wind turbines• Turbine design for large scale wind farm

Assess wind turbine RAMS aspectsin the context of the whole wind farm

21


Yearlyfailurefreq.

Present500 kW

DOWECtarget (5 MW)

Assumed1.5 - 2 MW

1

2

0

Wind turbine’s reliability


Reduction of failure frequencies

• Electric system 0.7 0.10• Blades 0.7 0.07 / 0.11• Yaw System 0.65 0.15• blade tips 0.5 0.14• Pitch Mechanism 0.5 0.13 / 0.14• Gearbox 0.5 0.13 / 0.15• Inverter 0.5 0.16• Control system 0.5 0.15 / 0.19

Components factor events/year

22


Base line

Active Stall

3 blades

2 speed

Tubular tower

Piled (tripod)

Up wind

Advanced

Pitch

3 blades

Var. speed(30%)

Tubular tower

Piled (tripod)

Up wind

Robust

Stall

2 blades

Fixed speed

Tubular tower

Monopile

Up wind

Stall-teeter

Stall

2 blades

Var. speed(full)

Truss tower

Gravity

Down wind(teetered hub)

Smart stall

Stall

3 blades

Var. speed(full)

Tubular tower

Piled (tripod)

Up wind

Direct drive

pitch

3 blades

Var. speed(full)

Tubular tower

Piled (tripod)

Up wind

DOWEC concepts


basic control concepts: yearly failure frequencies

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

Base Line Advanced Robust Stall-teeter

Smart-stall

Directdrive

Cum

ulat

ive

failu

re fr

eque

ncy

ControlInverterGearboxPitch Mechanismblade tipsYaw SystemBladeElectricBrakesGeneratorShaft & Bearings

DOWEC concepts: reliability

target

23


DOWEC concepts: failure classesbasic control concepts: yearly failure frequencies

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

Base Line Advanced Robust Stall-teeter

Smart-stall

Directdrive

Cum

ulat

ive

failu

re fr

eque

ncy

Cat. 4: Small orno parts, 24 hrs

Cat. 3: Smallparts, 48 hrs

Cat. 2: Largecomponentsinternal crane

Cat. 1: Heavycomponents,external crane

target


Numerical Monte Carlo SimulationQuantity of Spare Parts in Stock

0

5

10

15

20

25

1

1231

2461

3691

4921

6151

7381

8611

9841

1107

1

1230

1

1353

1

1476

1

1599

1

1722

1

time [hours]

quan

tity

Spare Part 1Spare Part 2Spare Part 3Spare Part 4

Accumulative Availability of Wind Farm against Time

96.00%

97.00%

98.00%

99.00%

100.00%

101.00%

1

941

1881

2821

3761

4701

5641

6581

7521

8461

9401

1034

1

1128

1

1222

1

1316

1

1410

1

1504

1

1598

1

1692

1

time [hours]

Availability of Turbines

93.00%

94.00%

95.00%

96.00%

97.00%

98.00%

99.00%

100.00%

101.00%

turbine 1turbine 2turbine 3turbine 4turbine 5turbine 6turbine 7turbine 8turbine 9turbine 10turbine 11turbine 12

• Site conditions(wind and waves)

• Failurescomponents of turbine, and wind farm

• Maintenance strategy- ships and crew- immediate/batch repair- overhaul- stock keeping

24


DOWEC 500 MW wind farm

Accessibility 80 %

Base L

ine

Advance

d

Robust

Stall-te

eter

Smart-st

all

Direct

drive

80.00%

85.00%

90.00%

95.00%

100.00%

1.60E+09

1.70E+09

1.80E+09

1.90E+09

2.00E+09

2.10E+09

Year

ly y

ield

[kW

h]

target


DOWEC 500 MW wind farmAccessibility 60 %

Base L

ine

Advance

d

Robust

Stall-te

eter

Smart-st

all

Direct

drive

80.00%

85.00%

90.00%

95.00%

100.00%

1.60E+09

1.70E+09

1.80E+09

1.90E+09

2.00E+09

2.10E+09

Year

ly y

ield

[kW

h]

target

25


Yearly O&M costs of wind farm [Euro]

€ 0€ 5,000,000

€ 10,000,000€ 15,000,000€ 20,000,000€ 25,000,000€ 30,000,000€ 35,000,000€ 40,000,000€ 45,000,000

Base Line

Advanc

ed

Robust

Stall-te

eter

Smart-s

tall

Direct

drive

Euro

Fixed costsTransport costsCrew costsCat 4: No partsCat 3: Small partsCat 2: Large comp.Cat 1: Heavy lift

DOWEC 500 MW wind farm


Context of the project

• Consortium of industries and institutes• Project: ~ 500 MW offshore wind farm• Location: North Sea > 12 mile zone • Concepts of 5 MW wind turbines• Turbine design for large scale wind farm

Develop optimal crew transport strategy

26


The Target

Develop an optimal O&M strategy for crew transport in the DOWEC offshore wind farm

• 80 * 6 MW wind turbines• 43 km off the Dutch coast (“NL7”)• 40 PM operations per year• 1.5 failure per year per turbine

(120 (averaged) CM operations per year) • 1 shift (12h) per 24 hours


Access systems considered

No

Access system

Significant wave height

[m]

Average (1-hour) wind speed

[m/s]

1 Fictitious 0.75 N.A. 2 Rubber boat, jump onto

ladder 1.5 10

3 Offshore Access System (OAS)

2 11.5

4 Offshore Access System + (optimistic assumption)

3 15

5 Helicopter NA 20

27


Wind & waves from the NESS/NEXT database

• North European Storm Study• consortium of oil companies

(“NL7“ data made available by Shell)

• “hindcast“ data• wind fields based on pressure

data• application of wave models• verification with measurements

• 3- hours interval; 30*30 km grid• 30 years; 9 years complete

(long term correlation of wind and waves)

+

‘NL7’ location


Variables in the NESS/NEXT database

Characteristic values for each 3-hour period:• V (1-hour) mean wind speed (m/s)

at 10 m height• θV wind direction (degrees)• Hs significant wave height (m)• Tz mean zero upcrossing period (s)• Θm wave direction (degrees)

9 years of consecutive data

28


From 3 D scatter plot

NEXT database: built-in (long term)correlation between wind and wave data

• ?• ⇒

Hs →

pdf→

Vw→

pdf→

Tz →

pdf→


To 2 D relations

V

Hs

29


NESS/NEXT database relation for NL7

V

Hs1 2 3 4 5

5

10

15

20

25

Zodiac

OAS+

Helicopter

OAS


Weather windows 1Example:• Uninterrupted time

intervals Hs < 1.25 m

Windows:• 6 hours• 12 hours• 24 hours• ...... hours

30


Weather windows 2

0

20

40

60

80

100

120

0 10 20 30 40 50 60

Weather window [hours]

Perc

enta

geHeli summerHeli winterOAS+ summerOAS summerOAS+ winterZodiac summerOAS winterZodiac winterFictitious summerFictitious winter

Weather Window [hours]

%


Availability of the DOWEC wind farm

0.0

20.0

40.0

60.0

80.0

100.0

120.0

0 20 40 60 80 100

Accessibility [%]

Ava

ilabi

lity

[%]

Accessibility [%]

Ava

ilabi

lity

[%]

HeliOAS+OAS

Zodiac

31


Immediate maintenance action

0102030405060708090

100

20 40 60 80 100

Accessibility [%]

Dire

ct a

ctio

n an

d co

mpl

tetio

n % 12 hours

24 hours 48 hours 168 hours 336 hours 12 hours 24 hours 48 hours168 hours336 hours

Accessibility [%]

[%]


Average waiting time

0

48

96

144

192

240

288

336

40 50 60 70 80 90 100

Accessibility [%]

Ave

rage

nr

of w

aitin

g ho

urs

12 hours 24 hours 48 hours168 hours336 hours

Accessibility [%]

Wai

ting

hour

s

32


CONTOFAX overall results

No Access system Accessibility [%]

Availability [%]

1 Fictitious 34 49 2 Rubber boat, jump onto

ladder 71 83

3 Offshore Access System 84 91 4 Offshore Access System+

(optimistic assumption) 95 95

5 Helicopter 100 96

DOWEC reference wind farm: one shift with two crews


DOWECcrew transport conclusions

• Rubber boat landing strategy not feasible

• OAS wind farm availability > 90%

• OAS+ availability > 95%

• OAS+ and heli comparable availability

• Average waiting time for short maintenanceactions (<48h) is limited (10 to 20 h)

33


O&M Conclusions• Present wind turbine reliability insufficient

• Certainly for wind farms at remote sites• Different maintenance strategy needed

• Opportunity based (flexible service intervals)• Condition based maintenance

• High impact of heavy lifting operations on costs• Large offshore wind farms need integrated

design• Wind turbines designed for marine maintenance

(and installation !!) operations• Special purpose O&M hardware

Offshore Wind Farm Design: Operation and Maintenance

Education

Transcript of Offshore Wind Farm Design: Operation and Maintenance