19
WINDTEST Kaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 1 VALIDATION OF WIND SPEED DISTURBANCES TO CUPS AT THE METEORLOCICAL MAST ON THE OFFSHORE PLATFORM FINO1 USING WIND-LIDAR MEASUREMENTS Authors: Detlef Kindler WINDTEST K.-W.-Koog GmbH Andy Oldroyd Oldbaum Services Ltd. DOWNVInD EU-Project FINO BMU-Project Title

VALIDATION OF WIND SPEED DISTURBANCES TO CUPS AT THE

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 1

VALIDATION OF WIND SPEED DISTURBANCES TO CUPS AT THE METEORLOCICAL MAST ON THE OFFSHORE PLATFORM FINO1 USING WIND-LIDAR MEASUREMENTSAuthors:

Detlef KindlerWINDTEST K.-W.-Koog GmbH

Andy OldroydOldbaum Services Ltd.

DOWNVInD EU-Project

FINO BMU-Project

Title

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 2

• MotivationUndisturbed wind measurements offshore

• North Sea research platforms FINO-1 & -3

• Assessment of a wind LiDAR system offshore on FINO-1

• Investigation of the distorted wind flow by the met. mast structure on the FINO-1 platform

using a wind LiDAR

• Application to future measurements on FINO-3

Outline

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 3

• Wind speed measurements offshore masts wind statistic, speed and directionwind speed profileenergy yield analysis

• High quality data at all heightsfor all wind directions

• Attention:wind flow distortion by mast structure itself (in certain wind directions)

Motivation

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 4

FINO 3 (2008)

FINO 1 (2003)

& Alpha VentusGerman offshore

test site (2008)

FINO Research Platforms

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 5

SKN=NW

-28m SKN

-58m SKN

+20,5m SKN

+103 SKN

Layout FINO 1 vs. FINO 3

FINO 1

+ 20,5 m

+ 103 m

- 28 m

SKN

-23 m

-50 m

+22 m+26 m

+4.6 m

+105 m

FINO 3

+ 21,5 m

+ 105 m

- 23 m

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 6

Main Deck

3

4

5

6

6 a

C

B

DContainer2

1Pile

WS

T, P, F, N, S, dT

T, F

WS, WD

WS

WS, WD

WS

7

8

9

10

10 a

Top

A

11 WS

WS

WS

WS, WD

WS, WD

T, P, F, N, dT

Measurement Levels

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 7

Hauptdeck

3

4

5

6

6 a

C

B

DContainer2

1Pfahl

7

8

9

10

10 a

Top

A

11

WS Cup 1 Hz

WD Vane 1Hz

WS/WD SONIC 10 Hz

Wind Sensors

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 8

FINO-1 Wind Flow Distortion

Square Mast Base

B

N

S

EW

075°

015°

255° 105°

165°195°

285°

345°

cup onmast top

030°

240°

cup onboom 60m

A

B

A

A

A

B

Disturbed Sectors

B

A

A

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 9

FINO-1 Wind LiDAR Assessment

Level 1

103 m

81 m

61 m

ComparisonComparisontop level & 2 boom levelstop level & 2 boom levels

Cups, disturbedCups, disturbed

LiDAR, undisturbedLiDAR, undisturbed Level 2

Level3

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 10

Height:103 (78) m

WS range:

2 to 23 m/s

Slope:m = 0.97

Regr. coefficient:

R2 = 0.99

WS Regression 103 m (Top)

Cup

LiD

AR

Undisturbed top sectors

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 11

Analysis Sector 15°-75°, 105°-165°, 195°-255°, 295°-345°

103 (78) m 81 (56) m 61 (36) m10-min-avg. values 1965 / / Slope "m" 0.97 / / Regr. Coeff "R2" 0.99 / /

103 (78) m 81 (56) m 61 (36) m10-min-avg. values 6005 2589 2749Slope "m" 0.98 0,97 0,98Regr. Coeff "R2" 0.99 0.99 1,00

30° to 90° and 180° to 240°

1st Period, 4 weeks

2nd Period, 14 weeks

Regression Results

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 12

Correction Procedure

• Use regression results from undisturbed sectorsto correct LiDAR wind speeds

• Use corrected LiDAR wind speeds as undisturbedequivalent of cup wind speeds for all wind directions,i.e. as well in disturbed sectors

• Derive correction factors in WD ranges, i.e. as well in disturbed sectors

…to be applied to the full time series

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 13

Top WS corrections 103 m

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 14

Boom WS Corrections 81 m

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 15

FINO 1: Square Base

Main Wind Dire

ction SW

Met Mast Base

FINO 3: Triangular Base

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 16

N

S

EW

Main Wind Direction SW

Mess-mast

Main Deck

Heli Deck

Mess-mast

FINO 3 – Mast & Boom Orientation

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 17

Pure Wind Speed Level

70 m - Level 7

Cup

Cup

Cup

N

S

EW

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 18

Summary

• Cup vs. LiDAR comparison campaign on FINO1 LiDAR proved high quality and good availabilityallowed an analysis of mast disturbance effects

• Heavy disturbances of wind flow from mast structureneed pre-cautions

- slim mast structure- triangular mast shape three boom directions- sufficient boom length

• Consequences for FINO3triangular shaped, slim met. mast layoutfinal boom layout still to be defined

• Outlook:LiDAR as stand alone offshore wind monitoring tool

on smaller support structures (maybe floating…)on existing offshore structures …

… e.g. like oil & gas platforms… research project

WINDTESTKaiser-Wilhelm-Koog GmbH EOW 2007 Berlin Slide No. 19

The End

Thanks