Met-Ocean and wind resource related studies for Six nearshore windfarms in Denmark

(1)

27 FEBRUARY 2015

SIX NEARSHORE WINDFARMS IN DENMARK

1

Met-Ocean and wind resource related studies for Six nearshore windfarms in Denmark

Thomas Gierlevsen (THGI), Project Director Henning Lauridsen (HGLN), Senior Project Manager Flemming Langhans (FLLS), Chief Specialist Jasmin Bejdic (JABC), Project Manager

(2)

Agenda

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Wind speed and uncertainty at sites JABC

10. Questions and Answers

(3)

› Pre-investigations

› Desk-studies by DHI

› Geophysical and bathymetric surveys by EGS & GEO

› Wave- and current measurements by DHI

› No on-site wind measurements

› Tentative park layouts by DTU wind

› EIA process - ongoing

Introduction

27 FEBRUARY 2015

3

(4)

Purpose of Met-ocean studies:

› To provide information and data on wind, waves, water levels, currents, ice etc.

(met-ocean data) at a level of detail which will sufficiently enable bidders to submit a qualified financial bid for design and construction of the OWF during the Danish Energy Agency's concession bidding process

› The met-ocean report has been prepared according to IEC 61400-3,

International standard. Windturbines - Part 3: Design requirements for offshore wind turbines.

Introduction

(5)

27 FEBRUARY 2015

5

Certification of Met-Ocean reports:

› Certification by DNV-GL

› Ensure general compliance to IEC 61400-3

› Reduce risk for concession bidders regarding met-ocean parameters

› Concessionary must take responsibility for met-ocean parameters for detailed design (Park layout not yet fixed)

Introduction

(6)

Wind resource related studies:

› No site specific met-masts or lidar

› Meso-scale modelling carried out to

establish virtual met-mast data at the sites

› Meso-scale model validated against available met-mast data

› The generated wind data and information shall enable concession bidders to conduct economic energy yield calculations to

qualified financial bids during the Danish Energy Agency's concession bidding process.

› Third party validation check by DEWI

Introduction

(7)

Introduction

Site description – Vesterhav Nord OWF

27 FEBRUARY 2015

7

› Location

› Approximately 4 km from the coastline and

› 6 km south-west of the Thyborøn

› Water depths

› 16 - 28 m

› Main characteristics:

› High wind site

› Rough wave climate

› Dynamic seabed (sand waves)

› No sea ice

(8)

Introduction

Site description – Vesterhav Nord OWF Wind Conditions

› DMI-HIRLAM (2003-2013)

› Top: +10 m MSL

› Bottom: +100 m MSL

(9)

Introduction

Site description – Vesterhav Nord OWF

27 FEBRUARY 2015

9

NW storm Dec 2013 ('Bodil)

Wave rose (2003-2013) Wind rose (2003-2013) +100m

(10)

Introduction

Site description – Vesterhav Nord OWF

Seabed morphology

› 1 km wide channel (erosional feature towards NW)

› 28 m deep

› Ongoing erosion

(10-20 cm/yr from 2012 to 2013)

› Sand waves (in channel and towards south)

› Up to 5.7 m height

› Migrate towards north

(10 m/yr from 2008 to 2013)

› "Seabed changes not of a magnitude that would be expected to have an effect on the met-ocean design conditions…"

› "… sandwaves should be considered in foundation design

› "… local hydrodynamic forcing may be affected by sandwaves migrating towards north."

(11)

Introduction

Site description – Vesterhav Syd OWF

27 FEBRUARY 2015

11

› Location:

› North-west of Ringkøbing Fjord

› Approximately 4 km from the coastline

› 7 km north-west of the Port of Hvide Sande

› Water depths: 15 to 26 m

› Primary characteristics:

› High wind site

› Energetic wave climate

› Dynamic seabed (sandwaves)

› No sea ice

(12)

Introduction

Site description – Vesterhav Syd OWF

Wind Conditions

› DMI-HIRLAM (2003-2013)

› Top: +10 m MSL

› Bottom: +100 m MSL

(13)

Introduction

Site description – Vesterhav Syd OWF

27 FEBRUARY 2015

13

NW storm Nov 2007

(14)

Introduction

Site description – Vesterhav Syd OWF

Seabed morphology

› Sand mining

› Up to 3 mill m³/yr in 578-AA until April 2014 by KDI

› KDI seeking approval to extend use of 578-AA

› KDI seeking approval in Nordsø Område 3 SW of VHS

› Smooth seabed with small sand waves

› Up to 1 m height of sand waves

› Migrating towards north

(5-15 m/yr from 2010 to 2013)

› "Seabed changes not of a magnitude that would be expected to have an effect on the met-ocean design conditions…"

› "… sandwaves should be considered in foundation design

› "… local hydrodynamic forcing may be affected by sandwaves migrating towards north."

(15)

Introduction

Site description – Sæby

27 FEBRUARY 2015

15

Sæby OWF

› Location:

› Between Nordjylland and Læsø

› 7 km south-west of the Port of Frederikshavn

› Water depths: 6 to 18 m

› Primary characteristics:

› Mild wind site

› Calm wave climate

› Gently sloping seabed

› Risk of sea ice

(16)

Introduction

Site description – Sæby OWF

Wind Conditions

› DMI-HIRLAM (2003-2013)

› Top: +10 m MSL

› Bottom: +100 m MSL

(17)

Introduction

Site description – Sæby OWF

27 FEBRUARY 2015

METOCEAN AND WIND RESOURCE-RELATED STUDIES

17

NE storm Nov 2006

(18)

Introduction

Site description – Sejerø Bugt OWF

› Location:

› Between Sjælland, Samsø and Sejerø

› 15 km north of the Port of Kalundborg

› Water depths: 7 to 22 m

› Primary characteristics:

› Intermediate wind site

› Intermediate wave climate

› Smooth seabed with channel of up to 24 m water depth

› Risk of sea ice

(19)

Introduction

Site description – Sejerø Bugt OWF

27 FEBRUARY 2015

19

Wind Conditions

› DMI-HIRLAM (2003-2013)

› Top: +10 m MSL

› Bottom: +100 m MSL

(20)

Introduction

Site description – Sejerø Bugt OWF

SW storm Oct 2013 ('Allan')

Wave rose (2003-2013)

Wind rose (2003-2013) +100m

(21)

Introduction

Site description – Smålandsfarvandet OWF

27 FEBRUARY 2015

21

› Location:

› Between Sjælland, Langeland and Lolland

› Approximately 6-10 km from the coastline

› 20 km south of the Port of Korsør

› Water depths: 3 to 20 m

› Primary characteristics:

› Calm wind site

› Calm wave climate

› Rough, stable seabed

› Risk of sea ice

(22)

Introduction

Site description – Smålandsfarvandet OWF

Wind Conditions

› DMI-HIRLAM (2003-2013)

› Top: +10 m MSL

› Bottom: +100 m MSL

(23)

Introduction

Site description – Smålandsfarvandet OWF

27 FEBRUARY 2015

23

SW storm Oct 2013 ('Allan')

Wave rose (2003-2013)

Wind rose (2003-2013) +100m

(24)

Introduction

Site description – Bornholm OWF

› Location:

› Southwest of Bornholm

› 8 km south of the Port of Rønne

› Water depths: 10 to 23 m

› Primary characteristics:

› Intermediate wind site

› Exposed wave climate

› Rough seabed with rock ridges and outcrops

› Risk of sea ice

(25)

Introduction

Site description – Bornholm OWF

27 FEBRUARY 2015

25

Wind Conditions

› DMI-HIRLAM (2003-2013)

› Top: +10 m MSL

› Bottom: +100 m MSL

(26)

Introduction

Site description – Bornholm OWF

W storm Dec 2013 ('Bodil') Wave rose (2003-2013)

Wind rose (2003-2013) +100m

(27)

Agenda

27 FEBRUARY 2015

27

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Wind speed and uncertainty at sites JABC

10. Questions and Answers

(28)

Wind Conditions

› Scope

› Air properties at 10m and 100m

› Turbulence and wind shear

› Normal wind speed at 10m and 100m

› Long-term average wind speed

› Extreme 10-minute average wind speed at 10m and 100m

› Methodology and Results

› DMI-HIRLAM model data (2003-2013)

› Validation

› Long-term correction

› Extreme value analysis

› Adjustment of extreme wind speed at for design

› Data on FTP (raw HIRLAM time series)

(29)

› HIRLAM =

High Resolution Limited Area Model

› State-of-the-art meteorological weather forecast model

› Data assimilation based on extensive amount of observational data of

atmospheric parameters

› 1 hourly boundary conditions from ECMWF

› Resolution: 15-5-3 km

DMI HIRLAM model

27 FEBRUARY 2015

29

(30)

› Extensive general and ongoing validation by DMI

› Site specific validation

against observed wind speed at 10m

› Challenge: No existing validation for elevations above 10m

DMI HIRLAM model validation

Wind Conditions

(31)

DMI HIRLAM model validation

Bornholm Airport (10m) during 'Bodil' storm

27 FEBRUARY 2015

31

(32)

DMI HIRLAM model validation

Arkona (10m) during 'Bodil' storm

Wind Conditions

(33)

DMI HIRLAM model validation

Arkona (10m), entire period 2003-2013

27 FEBRUARY 2015

33

(34)

› HIRLAM data (2003-2013) compared to the 35 years of WRF model data (1979-2013) used in the wind resource related study

› Long-term average wind speed based on the 35 years WRF data

Long-term validation of HIRLAM mean wind speed

Wind Conditions

(35)

Long-term Weibull mean wind speed at 100m

27 FEBRUARY 2015

35

9.6 9.9 9.9

9.1 9.3

9.5

(36)

› Extreme value analysis of DMI HIRLAM data

› Subsequent adjustments of recommended design value due to:

› The period 2003-2013 may not be

representative to the long-term extreme

wind statistics at the site. Increase of 0-15%

depending on the site.

› 'Disjunct sampling' effect. IEC 61400-3 based on extreme 10-min. average wind speed.

HIRLAM data presents 10-min. average wind speed every hour and not every 10 min.

Increase of 5%.

› HIRLAM underestimates the extreme wind speeds at 100m height. Increase of 4-13%

depending on the site

› Resulting increase of 13-20% depending on the site.

Extreme wind speed

Wind Conditions

(37)

27 FEBRUARY 2015

37

Scaling of HIRLAM extreme wind speeds at 100m

based on available met-mast data

(38)

Scaling of HIRLAM extreme wind speeds at 100m

Wind Conditions

(39)

27 FEBRUARY 2015

39

Scaling of HIRLAM extreme wind speeds at 100m

(40)

Recommended 50 year extreme 10-min. wind speeds at 100m

Wind Conditions

35 40 45 50

50 yearU10min(m/s)

41 44 46

40 38

40

(41)

Agenda

27 FEBRUARY 2015

41

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Wind speed and uncertainty at sites JABC

10. Questions and Answers

(42)

Water Levels

› Scope

› Normal water levels (tide and surge)

› Extreme water level (high and low)

› Methodology and Results

› Global Sea Level Rise & Land up-lift

› Tide and residual

› Analysis and derived parameters

› Data on FTP

(43)

27 FEBRUARY 2015

43

› Danish Coastal Directorate (KDI) recommendation:

› SLR (2100): 0.8 ±0.6 m

› SLR (2050): 0.3 ±0.2 m

› Lower in the northern part due to land uplift (rebound from last glaciation)

› Assumed SLR in 2045:

› Generally: 0.26m

› Sæby: 0.20m

Global Sea Level Rise (SLR)

Sea Level Rise in 2100

(44)

› Water level measurements from nearby stations

› Separation of tide and residual

› Extreme value analysis

› Comparison with KDI high

water statistics based on longer period of observations

Extreme water levels

Water Levels

Sæby / Frederikshavn

(45)

Extreme water levels

27 FEBRUARY 2015

45

1.42

-1.08

1.72

-1.54 1.93

-1.02 1.89

-0.85

2.97

-1.07

3.20

-1.71

(46)

Agenda

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Wind speed and uncertainty at sites JABC

10. Questions and Answers

(47)

27 FEBRUARY 2015

47

Wave Conditions

› Scope

› Wind - wave correlation

› Water level – wave correlation

› Normal sea state

› Normal wave height

› Extreme sea state

› Extreme wave height

› Methodology and Results

› MIKE 21 SW modelling

› Observations/measurements

› Validation

› Data on FTP

(48)

Wave hindcast modelling with MIKE 21 SW

Wave Conditions

› 11 year time series (2003 to 2013) of wave conditions at each site

› MIKE21 SW wave transformation modelling

› Boundary conditions from DMI-WAM regional wave model

› Wind fields from HIRLAM

› Varying water level from nearby

observations

Resolution: 1.85 km

(49)

Model set-up Vesterhav sites

27 FEBRUARY 2015

49

(50)

Wave Conditions

Fjaltring

Vesterhav Sites:

Validation against wave measurements at Fjaltring

(51)

Validation against wave measurements at Fjaltring

27 FEBRUARY 2015

51

(52)

Validation against wave measurements at Fjaltring

Wave Conditions

(53)

27 FEBRUARY 2015

53

Other Sites:

Validation against DHI site specific measurements

› Oct 2013 – Jan 2014

(54)

Validation against DHI measurements, Bornholm

Wave Conditions

(55)

Validation against DHI measurements, Bornholm

27 FEBRUARY 2015

55

(56)

Example of modelled wave conditions during storms Vesterhav Nord during 'Bodil' in Dec. 2013

Wave Conditions

(57)

Example of modelled wave conditions during storms Sejerø Bugt during 'Allan' in Oct 2013

27 FEBRUARY 2015

57

(58)

› Extreme value analysis significant wave height (H

_s

)

› Check for depth limitation:

, , 0.56 · ^.

Extreme Sea State (ESS)

Wave Conditions

(59)

50 year extreme significant wave heights

27 FEBRUARY 2015

59

0,0 2,0 4,0 6,0 8,0 10,0

Significant wave height HS(m)

(60)

› IEC 61400-3: Extreme individual wave height defined as:

1.86 · ,!""

› Check for depth limitation:

, 0.78 ·

› Extreme wave height determined for different water levels: LWL, MSL and HWL

› Wave period range:

$ _,!%& 0.80 · $_',

$ _(,!%& 1.05 · $_', ₍

› 50 year max. wave crest height and elevation determined from Stream function wave theory

Extreme Wave Height (EWH)

Wave Conditions

Extreme Wave Height (H_EWH)

Max. crest level

(61)

Agenda

27 FEBRUARY 2015

61

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Wind speed and uncertainty at sites JABC

10. Questions and Answers

(62)

Current Conditions

› Scope

› Normal currents

› Extreme currents

› Methodology and Results

› DMI-HBM model (2003-2013)

› MIKE 21 HD model

(Sejerø and Smålandsfarvandet)

› Validation and scaling

› Data on FTP

(raw and scaled HBM currents)

(63)

27 FEBRUARY 2015

63

› HBM = HIROMB-BOOS* Model

› DMI's operational 3D hydrodynamic model

› Resolution: 3-1-0.5 nm

› General ongoing validation by DMI against water level and current observations

› 11 years of DMI-HBM current data at all 6 sites (2003-2013)

DMI HBM model

* HIROMB = High Resolution Oceanographic Model for the Baltic.

BOOS = Baltic Operational Oceanographic System

(64)

Current Conditions

DMI HBM model validation

› Comparison with current speed measured by DHI during 'Bodil' storm in Dec. 2013

Sejerø Bugt

Smålandsfarvandet

(65)

27 FEBRUARY 2015

65

DMI-HBM model validation

DHI Measurements DMI-HBM

(66)

Current Conditions

Scaling of DMI-HBM current speeds

(67)

Detailed MIKE21 modelling for Sejerø and Smålandsfarvandet

27 FEBRUARY 2015

67

(68)

Detailed MIKE21 modelling, Sejerø Bugt

Current Conditions

(69)

Detailed MIKE21 modelling, Sejerø Bugt

27 FEBRUARY 2015

69

(70)

Detailed MIKE21 modelling, Sejerø Bugt

Current Conditions

(71)

'Area factors' for extreme currents speeds

27 FEBRUARY 2015

71

North-going currents South-going currents

(72)

50 year extreme depth-averaged current speeds

Current Conditions

(73)

Agenda

27 FEBRUARY 2015

73

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Wind speed and uncertainty at sites JABC

10. Questions and Answers

(74)

Ice Conditions

Henning Lauridsen

› Scope

› Extreme Ice Thickness

› Days with ice/heavy ice

› Methodology and Results

› Observations from Danish Ice Service

› Long term temperature measurements (>100 years)

› Ice charts

› Results

(75)

Observations and Measurements

27 FEBRUARY 2015

75

› Nautical-Meteorological Year-book by Danish Ice Service-SOK (1907-2014)

› Selected stations

› Max ice thickness per winter

› Number ice days and ice characteristics

(76)

Observations and Measurements

Ice Conditions

›

Nautical-Meteorological Year-book by Danish Ice Service-SOK (1907-2014)

› Average Temperature Measurements

› Tranebjerg (1862-2014)

› Vestervig (1872-2014)

› Sandvig (1874-2014)

› Bornholm Airport (1959-2013)

(77)

Observations and Measurements

27 FEBRUARY 2015

77

›

Nautical-Meteorological Year-book by Danish Ice Service-SOK (1907-2014)

›

Average Temperature Measurements

› Tranebjerg (1862-2014)

› Vestervig (1872-2014)

› Sandvig (1874-2014)

› Bornholm Airport (1959-2013)

› Accumulated winter sum of Freezing Days (AFDD)

› Danish Ice Service (1907-2014)

› Ice Charts

› Danish Ice Service (selected winters)

› German Ice Service (selected winters)

1942-03-31

(78)

WINTER [YEAR]

(BORNHOLM AFDD [DEG DAYS]OWF)

(DK ICE AFDD SERVICE) [DEG DAYS]

LIGHT [DAYS]ICE

HEAVY [DAYS]ICE

[DAYS]ICE

MAXIMUM ICE THICKNESS

[CM]

… … … … … … …

1933-34 7 9 0 0 0 0

1934-35 39 45 0 0 0 0

1935-36 35 36 0 0 0 0

1936-37 101 91 0 0 0 0

1937-38 20 15 0 0 0 0

1938-39 52 42 0 0 0 0

1939-40 349 306 53 19 72 information is missing

1940-41 229 209 22 0 22 15

1941-42 507 452 30 53 83 55

1942-43 55 53 0 0 0 0

1943-44 2 - 0 0 0 0

1944-45 57 54 0 0 0 0

1945-46 76 74 0 0 0 0

1946-47 377 340 13 49 62 information is missing

1947-48 52 50 0 0 0 0

… … … … … … …

Observations and Measurements

Ice Conditions

(79)

Analysis and derived parameters

27 FEBRUARY 2015

79

› AFDD and Ice thickness is proportional (DNV-OS-J101):

3.2 0.9,-.. − 50 [12]

› Global warming has a significant influence on the development of ice

› De-trending of AFDD to account for global warming.

1908-09 1911-12

1916-17 1921-22

1923-24 1928-29

1936-37

1939-40

1940-41

1941-42

1946-47

1953-54

1954-55

1955-56

1962-63

1969-70

1977-78

1978-79 1981-82 1984-85 1985-86

1986-87

1995-96

1996-97 2009-10 2010-11

0 10 20 30 40 50 60 70 80 90 100

0 100 200 300 400 500

[cm]

AFDD [Degree days]

Ice Thickness [cm] DNV (2013)

Sejerø Bugt

(80)

Analysis and derived parameters

› AFDD and Ice thickness is proportional (DNV-OS-J101):

3.2 0.9,-.. − 50 [12]

› Global warming has a significant influence on the development of ice

› De-trending of AFDD to account for global warming.

Ice Conditions

Tranebjerg

-2 -1 0 1 2 3 4 5

1873 1878 1883 1888 1893 1898 1903 1908 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013

Month average temperature [°C]

Winter

Linear (December) 10 per. Mov. Avg. (December) 10 per. Mov. Avg. (January) Linear (January)

Linear (February) 10 per. Mov. Avg. (February)

Linear (March) 10 per. Mov. Avg. (March)

Average Temperature Increase per 100 years [°C]

NOV DEC JAN FEB MAR 1.6 1.2 1.1 0.9 1.4

(81)

Analysis and derived parameters

27 FEBRUARY 2015

81

› AFDD and Ice thickness is proportional (DNV-OS-J101):

3.2 0.9,-.. − 50 [12]

› Global warming has a significant influence on the development of ice

› De-trending of AFDD to account for global warming.

Tranebjerg (AFDD)

Winter (RAW) 2015 1980 1950 1920

… … … …

1935-36 42 24 30 38 48

1936-37 77 50 62 72 84

1937-38 36 21 27 33 40

1938-39 61 40 49 58 67

1939-40 403 344 370 394 420 1940-41 316 258 285 309 334 1941-42 508 447 476 501 527

1942-43 58 41 48 56 65

1943-44 16 6 10 15 22

1944-45 60 43 51 58 68

1945-46 95 66 79 93 107

1946-47 417 359 388 414 441

1947-48 77 57 67 76 86

1948-49 28 16 22 27 34

1949-50 62 46 54 62 70

1950-51 52 27 39 53 70

… … … …

(82)

Analysis and derived parameters

› AFDD and Ice thickness is proportional (DNV-OS-J101):

3.2 0.9,-.. − 50 [12]

› Global warming has a significant influence on the development of ice

› De-trending of AFDD to account for global warming.

Ice Conditions

Tranebjerg (AFDD)

1941-42

1946-47 1939-40 1962-63

1940-41 1880-81 1928-29 1892-93 1955-56 1986-87 1978-79 1878-79 1969-70 1887-88 1981-82 1923-24 1984-85 1894-95 1885-86 1985-86 1900-01 1995-96 1890-91 1916-17 1899-00 1921-22 1957-58 1954-55 1965-66 1908-09

1941-42

1946-47 1939-40 1962-63

1940-41 1880-81 1928-29 1892-93 1955-56 1986-87 1978-79 1878-79 1969-70 1887-88 1981-82 1923-24 1984-85 1894-95 1885-86 1985-86 1900-01 1995-96 1890-91 1916-17 1899-00 1921-22 1957-58 1954-55 1965-66 1908-09

0 100 200 300 400 500 600

1,0 10,0 100,0 1000,0

AFDD [Degree days]

Return period [yr]

Corrected AFDD Raw AFDD

Log. (Corrected AFDD) Log. (Raw AFDD)

(83)

Analysis and derived parameters

Examples from Sejerø Bugt

27 FEBRUARY 2015

83

1906-071908-091916-17 1921-22

1928-29 1939-40

1940-41

1941-42 1946-47

1955-56

1962-63

1965-66

1978-79 1981-82 1984-85 1985-86

1986-87

1995-96 1906-07

1908-09

1911-12

1912-13

1916-17 1921-22

1923-24 1928-29

1936-37

1939-40

1940-41

1941-42

1946-47

1953-54

1954-55

1955-56

1962-63

1964-65 1965-66

1969-70 1978-79 1981-82 1984-85

1985-86 1986-87

1995-96

1996-972009-10 2010-11

0 10 20 30 40 50 60 70 80 90 100

0 100 200 300 400 500

[days]

AFDD (RAW) [Degree Days]

Heavy Ice [days] Ice [days]

Linear (Heavy Ice [days]) Linear (Ice [days])

1908-09 1911-12

1916-17 1921-22

1923-24 1928-29

1936-37

1939-40

1940-41

1941-42

1946-47

1953-54

1954-55

1955-56

1962-63

1969-70

1977-78

1978-79 1981-82 1984-85 1985-86

1986-87

1995-96

1996-97 2009-10 2010-11

0 10 20 30 40 50 60 70 80 90 100

0 100 200 300 400 500

[cm]

AFDD [Degree days]

Ice Thickness

No of ice days

(84)

1986-1987 Ice winter at Sejerø Bugt OWF

Ice Conditions

0. New ice or dark nilas (less than 5 cm thick).

1. Light nilas (5-10 cm thick).

2. Grey ice (10-15 cm thick).

3. Grey-white ice (15-30 cm thick).

4. White ice, first stage (30-50 cm thick).

5. White ice, second stage (50-70 cm thick).

6. Medium first year ice (70-120 cm thick).

(85)

› What is the quality of the historical measurements?

› Should the design ice thickness be based of an upper limit evaluation of measurements?

1986-1987 Ice winter at Sejerø Bugt OWF

27 FEBRUARY 2015

85

1908-09 1911-12

1916-17 1921-22

1923-24 1928-29

1936-37

1939-40

1940-41

1941-42

1946-47

1953-54

1954-55

1955-56

1962-63

1969-70

1977-78

1978-79 1981-82 1984-85 1985-86

1986-87

1995-96

1996-97 2009-10 2010-11

0 10 20 30 40 50 60 70 80 90 100

0 100 200 300 400 500

[cm]

AFDD [Degree days]

› AFDD and Ice thickness is proportional (DNV-OS-J101):

3.2 0.9,-.. − 50 [12]

› Of 25 winters with ice the expression

overestimates the ice thickness in 19

winters.

(86)

Results

Ice Conditions

(87)

Agenda

27 FEBRUARY 2015

87

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Wind speed and uncertainty at sites JABC

10. Questions and Answers

(88)

Questions?

(89)

Agenda

27 FEBRUARY 2015

89

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Site wind resource and uncertainty JABC

10. Questions and Answers

(90)

Mesoscale modelling and validation

Jasmin Bejdic

› Why mesoscale modelling?

› Methodology

› StormGeo WRF mesoscale model

› Modelling domains

› Validation analysis and derived parameters

› Result of validation

(91)

› No existing on-site measurements

› Political decision from the Danish Energy Agency in order to save on developing costs

› Not known which and how many of the six sites will be developed

› Best alternative when validated against high quality measurements

Why mesoscale modelling?

27 FEBRUARY 2015

91

(92)

› WRF = Weather Research and Forecasting

› State-of-the-art hindcast model with StormGeo improvments

› Boundary and initial data is ECMWF ERA-Interim

› Spatial resolution: 27-9-3-1 km

› Temporal resolution: 1 hour

› Simulation period: 01-09-2011 to 31-08-2013

StormGeo WRF mesoscale model

Mesoscale modelling and validation

(93)

Modelling domains

27 FEBRUARY 2015

93

(94)

Observations/measurements

› Identifying relevant met masts for mesoscale model validation

› Gathering met mast data and available met mast information

› Filtering data for errors, tower shadow effect

› Data analysis and uncertainty

estimation

(95)

27 FEBRUARY 2015

95

Observations/measurements

(96)

› Measurement uncertainties

Observations/measurements

1) Guidelines and recommendations from "Recommended practices for wind turbine testing – 11.

Wind speed measurement and use of cup anemometry. 1. Edition"

(97)

› Comparison of mesoscale data ("virtual time series") and validation met mast data

› Mean wind speed bias

› Comparison of frequency distributions

› Correlation analysis

› Wind rose comparison

› Wind shear comparison

› Comparison of Weibull A and k parameters

Validation analysis and derived parameters

27 FEBRUARY 2015

97

(98)

Validation analysis and derived parameters

› Example from Høvsøre mesoscale validation

› Mean wind speed bias is -0.1 m/s at 116 m

› 1 % model underestimation

compared to observed mean wind speed

0 5 10 15 20 25 30 35

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Wind Speed [m/s]

Cumulative Probability

Cumulative Density Function Høvsøre at 116 m

Observed Modelled

0 5 10 15 20 25 30 35

Wind Speed - Observed [m/s]

Wind Speed - Modelled [m/s]

Correlation - MOD versus OBS Høvsøre at 116 m

y = 0.88545x +1.043 R² = 0.77841 Data points

1. order linear regression

0 5 10 15 20 25 30 35

Observed Wind Speed [m/s]

Modelled Wind Speed [m/s]

Quantile-Quantile Plot [0:0.05:100] - Høvsøre at 116 m

2 4 6 8 10 12 14

0 20 40 60 80 100 120 140 160

Wind Speed [m/s]

Height [m]

Vertical Wind Profile - Modelled vs. Observed at Høvsøre

Modelled, α = 0.203 Observed, α = 0.182

(99)

Validation analysis and derived parameters

27 FEBRUARY 2015

99

› Example from Høvsøre mesoscale validation

› Wind roses

› Weibull parameters ^10%

15%

20%

Wind Rose - Modelled 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25 25 - 30 30 - 35

10%

15%

20%

Wind Rose - Observed 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25 25 - 30 30 - 35

0 5 10 15 20 25 30 35

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1

Wind speed [m/s]

Frequency of occurrence

Modelled and measured Weibull distributions at Høvsøre at 116 m A,obs = 11.39, k,obs = 2.4302 A,mod = 11.26, k,mod = 2.3716

(100)

› Validation uncertainties

Validation analysis and derived parameters

(101)

Results of validation

27 FEBRUARY 2015

101

› Mean wind speed bias within the combined uncertainty for most of the validation masts

› For all offshore and coastal met masts biases within or close to actual measurement uncertainty were found

› No bias to be applied to mesoscale wind data

› The combined uncertainty shall be

"transferred" to each project site

(102)

Agenda

Introduction

1. Purpose, Site-descriptions THGI Met-ocean conditions

2. Wind conditions THGI

3. Water levels THGI

4. Wave conditions THGI

5. Current conditions THGI

6. Sea ice HGLN

7. Questions and Answers Wind resource related studies

8. Mesoscale modelling and validation JABC

9. Site wind resource and uncertainty JABC

10. Questions and Answers

(103)

27 FEBRUARY 2015

103

Site wind speed and uncertainty

› Methodology

› Transferring validation mast uncertainty to project sites

› Wind resource

› Virtual met masts

› Results

› Deliverables

› Independent third party validation

(104)

Transferring validation mast uncertainty to project sites

Site wind resource and uncertainty

(105)

› Uncertainty contribution from every validation mast

› Masts of higher relevance is given a higher weighting according to:

› Mast height

› Terrain type

› Source data length and temporal resolution

› Distance to project site

Transferring validation mast uncertainty to project sites

27 FEBRUARY 2015

105

(106)

› Qualitative assessment

› Base uncertainty obtained from weighting process was subject to a qualitative assessment and scaled according to the project site location

› Generally higher uncertainty at the east coast of Jutland and inner Danish waters

› The transferred uncertainty represents the uncertainty of the generated site- specific wind data

› T

Transferring validation mast uncertainty to project sites

The estimated uncertainty is given as the standard

uncertainty corresponding to a

confidence level of 68 per cent

(107)

› 2-year period at 100 m

Wind resource

27 FEBRUARY 2015

107

(108)

Virtual met masts

(109)

Results

27 FEBRUARY 2015

109

Site Statistic

s P1 P2 P3 P4 Uncertainty

[%]

Vesterhav Syd

V_mean

[m/s] 9.91 9.98 9.98 9.86 3.3

A [m/s] 11.26 11.32 11.33 11.21 -

k 2.367 2.367 2.375 2.371 -

Vesterhav Nord

V_mean

[m/s] 9.98 10.02 10.02 10.09 3.2

A [m/s] 11.32 11.38 11.36 11.46 -

k 2.238 2.323 2.291 2.321 -

Sæby

V_mean

[m/s] 9.11 9.03 9.15 9.17 4.6

A [m/s] 10.37 10.27 10.41 10.44 -

k 2.274 2.254 2.249 2.268 -

Sejerø

V_mean

[m/s] 9.45 9.48 9.39 9.40 4.6

A [m/s] 10.83 10.89 10.77 10.78 -

k 2.441 2.427 2.412 2.421 -

Smålandsfarvandet

V_mean

[m/s] 9.23 9.24 9.30 9.24 5.1

A [m/s] 10.60 10.62 10.68 10.61 -

k 2.443 2.440 2.444 2.455 -

Bornholm

V_mean

[m/s] 9.51 9.38 9.52 9.55 3.4

A [m/s] 10.79 10.62 10.80 10.84 -

k 2.289 2.235 2.282 2.301 -

› 2-year period

(110)

Long-term corrected mean wind speed at 100m

9,5

9,3 9,2

9,0

9,8 9,8

8,6 8,8 9 9,2 9,4 9,6 9,8 10

MEAN WIND SPEED (M/S)

(111)

› Wind roses

Results

27 FEBRUARY 2015

111

10%

15%

Wind Rose - VHN1 at 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25 25 - 30 30 - 35 35 - 40

10%

15%

Wind Rose - VHS1 at 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25 25 - 30 30 - 35

10%

15%

Wind Rose - S1 at 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25 25 - 30

>=30

Sejerø

10%

15%

Wind Rose - SØ1 at 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25 25 - 30

Sæby

10%

15%

Wind Rose - SM1 at 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25 25 - 30

Smålandsfarvandet

10%

15%

Wind Rose - B1 at 100m

WEST EAST

SOUTH NORTH

0 - 5 5 - 10 10 - 15 15 - 20 20 - 25

>=25

Bornholm Vesterhav Syd

Vesterhav Nord

(112)

› Four "virtual wind measurement" points at each site consisting of

› 2-year time series of wind speed, wind direction, temperature and air pressure in 1- hour resolution

› Direct output from the WRF modelling

› Time series of the same parameters except air pressure for 34 years in 3-hour resolution from ERA-Interim

› 2-year time series long-term corrected using 34 years ERA-Interim data (Generalized Extreme Value Method)

Deliverables

(113)

› DEWI has validated the COWI approach and made completely independent calculations in order to verify the COWI derived uncertainties and results

Independent third party Validation

27 FEBRUARY 2015

113