NRELOffshrBsline5MW.

doc
Jason Jonkman, NREL/NWTC
June 14, 2006
Introduction:
This manuscript documents NRELs baseline wind turbine aeroelastic model for use in various
offshore analysis concept studies. This includes the aeroelastic FAST5 model, baseline control
system (torque and pitch), and rationale behind the model. The turbine parameters are fictitious
and do not represent an actual wind turbine, but do represent a good approximation of what an
actual wind turbine of the associated size would be like.

The FAST model can be used to generate an ADAMS7 model using the FAST-to-ADAMS
preprocessor built-into FAST. NREL has compared the responses between the FAST and
ADAMS model results for a number of simulationsthe results agree pretty well in general
the most notable differences are caused by the availability of mass offsets and blade torsion
DOFs in ADAMS, which are not available in FAST.

Numerous additional details and rationale may be found in the MS Excel workbook,
NRELOffshrBsline5MW.xls, which is available from:

Jason Jonkman
National Wind Technology Center (NWTC)
National Renewable Energy Laboratory (NREL)
1617 Cole Boulevard
Golden, CO 80401-3393
United States of America
Phone: +1 (303) 384 - 7026
Fax: +1 (303) 384 - 6901
E-mail: [email protected]

General Machine Properties:

Before creating the aeroelastic model, the size/rating had to be settled on. It was understood
from the onset that for a deepwater system to be cost-effective, each individual wind turbine
must be rated at 5MW or higher. Ratings considered as the baseline included 5MW, 6MW,
7MW 10MW and up to 20MW. It was decided that the baseline should be 5MW since:
Feasible floater configurations scoped out by NREL/NWTC were based on the
assumption of a 5MW unit13.
Cost studies at DOE were based on a rotor diameter of 128m13.
5MW has been used in a previous study, RECommendations for design of OFFshore
wind turbines (RECOFF), from Europe19.
The largest wind turbine prototypes in the world, the Multibrid M50001,11,12 and
REpower 5M9,16,17, currently have ratings of 5MW.
Machines, will likely get larger, not smaller.
Thus, 5MW is an appropriate size and has a broad base of industry precedence.

Next, the aeroelastic properties of the 5MW baseline had to be established. In order for the
concept studies to be defendable, the aeroelastic model had to be realistic. In order to achieve
realism, NREL gathered publicly available gross properties of several aeroelastic models and

1
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
created a composite from these systems, extracting the best available information. A summary
of the aeroelastic models utilized is given below:
FAST model of Global Energy Concepts (GECs) Wind Partnerships for Advanced
Component Development (WindPACT)14 5A0210.
FAST model of GECs WindPACT 5A0410.
BLADED model used during Garrad Hassans RECOFF project19.
BLADED and PHATAS models used during TU Delfts Dutch Offshore Wind Energy
Converter (DOWEC) project2,6,8.
and of the two largest wind turbine prototypes in the world:
Multibrid M50001,11,12
REpower 5M9,16,17
These properties are provided in the following table:

Machine Data Comparisons of Several 5MW Wind Turbines and Models

Parameter Units Machine Data*
Company (-) GEC GEC Garrad Hassan TU Delft Multibrid REpower
Project (-) WindPACT 5A02WindPACT 5A04RECOFF DOWEC M5000 5M
Fictitious/Real (-) Fictitious Fictitious Fictitious Fictitious Real Real
IEC Class (-) 1A 2A 1C 1A
Rated Power MW 5.000 5.000 5.000 6.000 5.000 5.000
Number of Blades (-) 3 3 3 3 3 3
Rated Wind Speed m/s 12.000 12.000 13.000
Diameter m 120.000 128.000 118.000 129.000 116.000 126.000
Blade Length m 57.000 60.800 57.000 62.700 61.500
Hub Diameter m 6.000 6.400 4.000 3.600 3.000
Hub Height m 100.000 100.000 80.000 91.400 85.000 90.000-100.000
Gearbox Ratio (-) 174.000 160.850 86.390 92.873 99.234 97.000
Rated Generator Speed rpm 1,800.000 1,800.000 1,200.000 1,100.000 1,468.663 1,173.700
Rated Rotor Speed rpm 10.345 11.191 13.890 11.844 14.800 12.100
Rated Tip Speed m/s 64.998 75.000 85.822 80.000 89.891 79.828
Overhang m 6.000 6.000 6.000 5.000
Blade Precone deg 0.000 0.000 0.000 2.500 2.000 >0
Shaft Tilt deg 5.000 5.000 4.000 5.000 5.000 >0
Blade Mass kg 21,170.000 27,812.000 13,813.000 17,905.000 16,500.000 17,740.000
Blade Second Mass Moment kg*m^2 11,078,654.000 8,004,681.000 11,676,630.000
Hub Mass kg 109,345.000 125,970.000 50,000.000 30,000.000 58,200.000 56,780.000
Hub Inertia kg*m^2 711,411.000 935,700.000 110,000.000 50,700.000
Rotor Mass kg 172,855.000 209,406.000 91,439.000 83,715.000 107,700.000 110,000.000
Rotor Inertia kg*m^2 41,676,612.000 27,392,980.000 39,047,480.000
Nacelle Mass kg 246,626.000 270,669.000 250,000.000 187,952.000 194,090.000 240,000.000
Nacelle Yaw Inertia kg*m^2 816,681.000 0.000 2,420,000.000
Tower Top Mass kg 419,481.000 480,075.000 341,439.000 271,667.000 301,790.000 350,000.000
Tower Mass kg 499,635.000 503,000.000 540,294.500 345,370.600
Total Mass of Turbine kg 919,116.000 983,075.000 881,733.500 617,037.600
Maximum Chord m 5.295 5.120 4.440 4.620 4.600
Maximum Twist deg 11.100 11.100 13.000 13.500

*Values highlighted in green are the closest match to REpower 5M

*Values highlighted in red are untuned estimates

Note that the rotor diameters supplied in the table above ignore the effects of blade precone,
which reduce the actual rotor diameter and swept area.

The Multibrid M5000 machine has a significantly higher tip speed than typical onshore wind
turbines and lower tower top mass than would be expected from recently developed scaling
laws18. In contrast, the REpower 5M machine has properties that are more expected. In this
sense, the REpower 5M machine is more conventional than the Multibrid M5000, so it makes

2
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
sense to use the specifications of the REpower 5M machine as the target specifications for the
model. Thus the task was to establish an aeroelastic model that tries to resemble the REpower
5M machine as close as possible (i.e., as close as publicly known) and to fill in the data gaps
using machine properties known from the DOWEC, RECOFF, and WindPACT studies. From
the table above, the DOWEC turbine most closely matched the REpower 5M machine, which
makes sense since the DOWEC study used aeroelastic blade properties provided by LM
glasfiber9, and the REpower 5M machine also utilizes LM glasfiber blades. As such, must of the
data gaps are filled in using machine properties from the DOWEC study6.

Knowing that the rotor radius for our aeroelastic model would be about 63m (from the REpower
5M machine), and wanting the lowest reasonable hub-height possible in order to minimize the
overturning moment of the platform, NREL decided that the hub height for the baseline model
should be 90m, which would give a 15m air gap between the blade tips at their lowest point and
a 30m (15m amplitude) 50-yr extreme wave height.

Additional gross properties chosen for the NREL 5MW baseline wind turbine, most of which are
identical to the REpower 5M, are provided in the table below:

Gross Properties Chosen for the NREL 5MW Baseline Wind Turbine Model
Rating 5MW
Wind Regime IEC 61400-3 (Offshore) Class 1B / Class 6 winds
Rotor Orientation Upwind
Control Variable Speed, Collective Pitch
Rotor Diameter / Hub Diameter 126m / 3m
Hub Height 90m
Maximum Rotor / Generator Speed 12.1rpm / 1,173.7rpm
Maximum Tip Speed 80m/s
Overhang / Shaft Tilt / Precone 5m / 5 / 2.5
Rotor Mass 110,000kg Overall c.g. location:
Nacelle Mass 240,000kg (x ,y ,z ) = (-0.2m,0.0m,64.0m)
t t t
Tower Mass (Deepwater) 347,460kg
Reference Site National Data Buoy Center (NDBC) Buoy 44008

The tower mass listed in the table above is for the deepwater support structure configuration.
The overall c.g. location of this wind turbine as indicated above is in the tower base coordinate
system whose origin lies on the tower centerline at the mean sea level (MSL).

The aeroelastic FAST model, which cant support prebend, incorporates a 2.5 upwind precone.
This is used to represent the small amount of precone and larger amount of precurve built-into
the actual REpower 5M, 61.5m long, LM glasfiber blade9. The rotor and hub diameters
indicated in the table above ignore the effects of blade precone, which reduces the actual
diameters and swept area. The actual rotor diameter in the model (assuming that the blades are
undeflected) is (126m)*COS(2.5) = 125.88m and the actual swept area is (/4)*(125.88m)^2 =
12,445.3m^2.

3
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
Blade Structural Properties:
The distributed blade structural properties of the NREL 5MW baseline model are based on the
structural properties of the 62.6m long LM Glasfiber blade used in the DOWEC study, as given
in Appendix A of Ref. 8. Since the blades in the DOWEC study are 1.1m longer than the
REpower 5M blades, the structural properties of the NREL 5MW baseline blade are found by
truncating the 62.6m blades at 61.5m blade span (the structural properties of the blade tip are
found by interpolating between the 61.2m and 61.7m stations provided in Appendix A of Ref. 8).
The resulting properties are given in the table below:

Distributed Blade Structural Properties

Radius BlFract AeroCent StrcTwst BMassDen FlpStff EdgStff GJStff EAStff Alpha FlpIner EdgIner PrecrvRef PreswpRef FlpcgOf EdgcgOf FlpEAOf EdgEAOf
(m) (-) (-) (deg) (kg/m) (Nm^2) (Nm^2) (Nm^2) (N) (-) (kg m) (kg m) (m) (m) (m) (m) (m) (m)
1.50 0.00000 0.25000 13.308 678.935 18110.00E+6 18113.60E+6 5564.40E+6 13942.85E+6 0.0 1394.03 1394.54 0.0 0.0 0.0 0.00017 0.0 0.0
1.70 0.00325 0.25000 13.308 678.935 18110.00E+6 18113.60E+6 5564.40E+6 13942.85E+6 0.0 1394.03 1394.54 0.0 0.0 0.0 0.00017 0.0 0.0
2.70 0.01951 0.24951 13.308 773.363 19424.90E+6 19558.60E+6 5431.59E+6 15054.91E+6 0.0 1540.56 1470.42 0.0 0.0 0.0 -0.02309 0.0 0.0
3.70 0.03577 0.24510 13.308 740.550 17455.90E+6 19497.80E+6 4993.98E+6 13707.94E+6 0.0 1260.31 1481.28 0.0 0.0 0.0 0.00344 0.0 0.0
4.70 0.05203 0.23284 13.308 740.042 15287.40E+6 19788.80E+6 4666.59E+6 13330.23E+6 0.0 1031.75 1634.30 0.0 0.0 0.0 0.04345 0.0 0.0
5.70 0.06829 0.22059 13.308 592.496 10782.40E+6 14858.50E+6 3474.71E+6 9981.35E+6 0.0 709.92 1286.36 0.0 0.0 0.0 0.05893 0.0 0.0
6.70 0.08455 0.20833 13.308 450.275 7229.72E+6 10220.60E+6 2323.54E+6 6886.27E+6 0.0 463.33 913.92 0.0 0.0 0.0 0.06494 0.0 0.0
7.70 0.10081 0.19608 13.308 424.054 6309.54E+6 9144.70E+6 1907.87E+6 6048.03E+6 0.0 378.31 831.29 0.0 0.0 0.0 0.07718 0.0 0.0
8.70 0.11707 0.18382 13.308 400.638 5528.36E+6 8063.16E+6 1570.36E+6 5279.51E+6 0.0 308.59 747.31 0.0 0.0 0.0 0.08394 0.0 0.0
9.70 0.13335 0.17156 13.308 382.062 4980.06E+6 6884.44E+6 1158.26E+6 4460.74E+6 0.0 261.56 630.59 0.0 0.0 0.0 0.10174 0.0 0.0
10.70 0.14959 0.15931 13.308 399.655 4936.84E+6 7009.18E+6 1002.12E+6 4329.30E+6 0.0 230.63 635.23 0.0 0.0 0.0 0.10758 0.0 0.0
11.70 0.16585 0.14706 13.308 426.321 4691.66E+6 7167.68E+6 855.90E+6 4460.98E+6 0.0 196.24 695.96 0.0 0.0 0.0 0.15829 0.0 0.0
12.70 0.18211 0.13481 13.181 416.820 3949.46E+6 7271.66E+6 672.27E+6 4625.25E+6 0.0 145.86 779.19 0.0 0.0 0.0 0.22235 0.0 0.0
13.70 0.19837 0.12500 12.848 406.186 3386.52E+6 7081.70E+6 547.49E+6 5024.26E+6 0.0 114.77 890.08 0.0 0.0 0.0 0.30756 0.0 0.0
14.70 0.21465 0.12500 12.192 381.420 2933.74E+6 6244.53E+6 448.84E+6 4374.89E+6 0.0 91.79 783.19 0.0 0.0 0.0 0.30386 0.0 0.0
15.70 0.23089 0.12500 11.561 352.822 2568.96E+6 5048.96E+6 335.92E+6 3479.90E+6 0.0 73.33 622.65 0.0 0.0 0.0 0.26519 0.0 0.0
16.70 0.24715 0.12500 11.072 349.477 2388.65E+6 4948.49E+6 311.35E+6 3261.55E+6 0.0 63.15 589.16 0.0 0.0 0.0 0.25941 0.0 0.0
17.70 0.26341 0.12500 10.792 346.538 2271.99E+6 4808.02E+6 291.94E+6 3029.59E+6 0.0 56.81 549.11 0.0 0.0 0.0 0.25007 0.0 0.0
19.70 0.29595 0.12500 10.232 339.333 2050.05E+6 4501.40E+6 261.00E+6 2564.99E+6 0.0 45.76 467.24 0.0 0.0 0.0 0.23155 0.0 0.0
21.70 0.32846 0.12500 9.672 330.004 1828.25E+6 4244.07E+6 228.82E+6 2167.76E+6 0.0 36.28 397.27 0.0 0.0 0.0 0.20382 0.0 0.0
23.70 0.36098 0.12500 9.110 321.990 1588.71E+6 3995.28E+6 200.75E+6 1881.07E+6 0.0 27.65 348.56 0.0 0.0 0.0 0.19934 0.0 0.0
25.70 0.39350 0.12500 8.534 313.820 1361.93E+6 3750.76E+6 174.38E+6 1619.67E+6 0.0 20.52 303.42 0.0 0.0 0.0 0.19323 0.0 0.0
27.70 0.42602 0.12500 7.932 294.734 1102.38E+6 3447.14E+6 144.47E+6 1247.78E+6 0.0 13.33 236.23 0.0 0.0 0.0 0.14994 0.0 0.0
29.70 0.45855 0.12500 7.321 287.120 875.80E+6 3139.07E+6 119.98E+6 1021.15E+6 0.0 8.51 195.72 0.0 0.0 0.0 0.15421 0.0 0.0
31.70 0.49106 0.12500 6.711 263.343 681.30E+6 2734.24E+6 81.19E+6 758.93E+6 0.0 5.30 146.49 0.0 0.0 0.0 0.13252 0.0 0.0
33.70 0.52358 0.12500 6.122 253.207 534.72E+6 2554.87E+6 69.09E+6 658.99E+6 0.0 3.35 128.44 0.0 0.0 0.0 0.13313 0.0 0.0
35.70 0.55610 0.12500 5.546 241.666 408.90E+6 2334.03E+6 57.45E+6 556.00E+6 0.0 2.03 109.17 0.0 0.0 0.0 0.14035 0.0 0.0
37.70 0.58862 0.12500 4.971 220.638 314.54E+6 1828.73E+6 45.92E+6 418.78E+6 0.0 1.36 82.39 0.0 0.0 0.0 0.13950 0.0 0.0
39.70 0.62115 0.12500 4.401 200.293 238.63E+6 1584.10E+6 35.98E+6 342.08E+6 0.0 0.88 67.53 0.0 0.0 0.0 0.15134 0.0 0.0
41.70 0.65366 0.12500 3.834 179.404 175.88E+6 1323.36E+6 27.44E+6 270.31E+6 0.0 0.55 53.51 0.0 0.0 0.0 0.17418 0.0 0.0
43.70 0.68618 0.12500 3.332 165.094 126.01E+6 1183.68E+6 20.90E+6 298.01E+6 0.0 0.32 59.28 0.0 0.0 0.0 0.24922 0.0 0.0
45.70 0.71870 0.12500 2.890 154.411 107.26E+6 1020.16E+6 18.54E+6 239.54E+6 0.0 0.25 47.66 0.0 0.0 0.0 0.26022 0.0 0.0
47.70 0.75122 0.12500 2.503 138.935 90.88E+6 797.81E+6 16.28E+6 177.13E+6 0.0 0.19 35.23 0.0 0.0 0.0 0.22554 0.0 0.0
49.70 0.78376 0.12500 2.116 129.555 76.31E+6 709.61E+6 14.53E+6 146.30E+6 0.0 0.14 29.11 0.0 0.0 0.0 0.22795 0.0 0.0
51.70 0.81626 0.12500 1.730 107.264 61.05E+6 518.19E+6 9.07E+6 96.79E+6 0.0 0.11 19.25 0.0 0.0 0.0 0.20600 0.0 0.0
53.70 0.84878 0.12500 1.342 98.776 49.48E+6 454.87E+6 8.06E+6 79.64E+6 0.0 0.08 15.85 0.0 0.0 0.0 0.21662 0.0 0.0
55.70 0.88130 0.12500 0.954 90.248 39.36E+6 395.12E+6 7.08E+6 64.66E+6 0.0 0.05 12.88 0.0 0.0 0.0 0.22784 0.0 0.0
56.70 0.89756 0.12500 0.760 83.001 34.67E+6 353.72E+6 6.09E+6 54.86E+6 0.0 0.05 10.93 0.0 0.0 0.0 0.23124 0.0 0.0
57.70 0.91382 0.12500 0.574 72.906 30.41E+6 304.73E+6 5.75E+6 28.04E+6 0.0 0.04 5.57 0.0 0.0 0.0 0.14826 0.0 0.0
58.70 0.93008 0.12500 0.404 68.772 26.52E+6 281.42E+6 5.33E+6 25.11E+6 0.0 0.03 4.99 0.0 0.0 0.0 0.15346 0.0 0.0
59.20 0.93821 0.12500 0.319 66.264 23.84E+6 261.71E+6 4.94E+6 22.22E+6 0.0 0.03 4.42 0.0 0.0 0.0 0.15382 0.0 0.0
59.70 0.94636 0.12500 0.253 59.340 19.63E+6 158.81E+6 4.24E+6 11.33E+6 0.0 0.02 2.25 0.0 0.0 0.0 0.09470 0.0 0.0
60.20 0.95447 0.12500 0.216 55.914 16.00E+6 137.88E+6 3.66E+6 8.61E+6 0.0 0.01 1.71 0.0 0.0 0.0 0.09018 0.0 0.0
60.70 0.96260 0.12500 0.178 52.484 12.83E+6 118.79E+6 3.13E+6 6.44E+6 0.0 0.01 1.28 0.0 0.0 0.0 0.08561 0.0 0.0
61.20 0.97073 0.12500 0.140 49.114 10.08E+6 101.63E+6 2.64E+6 4.77E+6 0.0 0.01 0.95 0.0 0.0 0.0 0.08035 0.0 0.0
61.70 0.97886 0.12500 0.101 45.818 7.55E+6 85.07E+6 2.17E+6 3.40E+6 0.0 0.00 0.68 0.0 0.0 0.0 0.07096 0.0 0.0
62.20 0.98699 0.12500 0.062 41.669 4.60E+6 64.26E+6 1.58E+6 1.94E+6 0.0 0.00 0.39 0.0 0.0 0.0 0.05424 0.0 0.0
62.70 0.99512 0.12500 0.023 11.453 .25E+6 6.61E+6 .25E+6 .38E+6 0.0 0.00 0.08 0.0 0.0 0.0 0.05387 0.0 0.0
63.00 1.00000 0.12500 0.000 10.319 .17E+6 5.01E+6 .19E+6 .23E+6 0.0 0.00 0.05 0.0 0.0 0.0 0.05181 0.0 0.0

The entries in the first column, Radius, are the spanwise locations along the blade pitch axis
relative to the rotor center (apex).

BlFract is the fractional distance along the blade pitch axis from the root (0.0) to the tip (1.0).
The blade root is located 1.5m along the pitch axis from the rotor center (apex).

AeroCent is a FAST variable name. By definition, the quantity (AeroCent - 0.25) is the
fractional distance to the aerodynamic center from the pitch axis along the chordline, positive
towards the trailing edge. Thus, at the root (BlFract = 0.0), AeroCent = 0.25 means that the
aerodynamic center lies on the pitch axis [since (0.25 - 0.25) = 0.0]. And at the tip (BlFract =
1.0), AeroCent = 0.125 means that the aerodynamic center lies 0.125 chordlengths toward the
leading edge from pitch axis [since (0.125 - 0.25) = -0.125].

4
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
The flapwise and edgwise section stiffness and inertia values, FlpStff, EdgStff, FlpIner and
EdgIner, are given about the principal structural axes of each cross section as oriented by the
structural twist angle, StrcTwst. The values of StrcTwst as given in the table above, are offset by
-0.09182 from the values provided in Appendix A of Ref. 8 in order to insure that the zero-twist
reference location is at the blade tip.

GJStff are the values of the blade torsional stiffness. Since the DOWEC blade data did not
contain extensional stiffness information, the blade extensional stiffness values, EAStff, given in
the table above were estimated to be 10^7 times the average mass moment of inertia at each
blade station (this rule of thumb comes from WindPACT data). This rule of thumb should get us
in the ballpark at leastthe exact values are not important due to the low rotational speed.

The flapwise c.g. offset values, FlpcgOf, and flapwise and edgewise elastic offset values,
FlpEAOf and EdgEAOf, given in the table above were assumed to be zero due to the inability to
decipher the corresponding information given in Appendix A of Ref. 8. The edgewise c.g. offset
values, EdgcgOf, are the distances in meters along the chordline from the pitch axis to the center
of gravity of the blade section, positive towards the trailing edge.

The distributed blade section mass per unit length values, BMassDen, given in the table above
are increased by 4.536% in the model in order to scale the overall (integrated) blade mass to
17,740kg, which is the mass of the blades in the REpower 5M machine. In the model, the
second mass moment of inertia, first mass moment of inertia, and the radial c.g. location of each
blade is 11,776,047 kg-m^2, 363,231 kg-m, and 20.475m with respect to the blade root,
respectively.

The NREL 5MW baseline model incorporates 0.477465% critical structural damping in all
modes of the isolated blade, which corresponds to the 3% logarithmic decrement used in the
DOWEC study from page 20 of Ref. 6.

The table below summarizes the undistributed blade structural properties discussed in this
section:

Undistributed Blade Structural Properties

Length (w.r.t. Root Along Preconed Axis) 61.5 m
Mass Scaling Factor 4.536 %
Overall (Integrated) Mass 17,740 kg
Second Mass Moment of Inertia (w.r.t. Root) 11,776,047 kg-m^2
First Mass Moment of Inertia (w.r.t. Root) 363,231 kg-m
c.g. Location (w.r.t. Root Along Preconed Axis) 20.475 m
Structural Damping Ratio (All Modes) 0.477465 %

Blade Aerodynamic Properties:

Like the blade structural properties, the blade aerodynamic properties are also based on the
DOWEC blade as described in Table 1 on page 13 of Ref. 6 and Appendix A of Ref. 8. The
model uses 17 blade elements for integration of the aerodynamic and structural forces. The
inboard 3 and outboard 3 elements are two-thirds the size of the 11 equally-spaced midspan
elements in order to better capture the large structural gradients at the blade root and the large

5
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
aerodynamic gradients at the blade tip. The aerodynamic properties at the blade nodes, which
are located at the center of the blade elements, are given in the table below. The blade node
locations, labeled as RNodes in the table, are directed along the pitch axis from the rotor center
(apex) to the blade cross sections.

Distributed Blade Aerodynamic Properties

Node RNodes AeroTwst DRNodes Chord Airfoil Table
(-) (m) (deg) (m) (m) (-)
1 2.8667 13.308 2.7333 3.542 Cylinder1.dat
2 5.6000 13.308 2.7333 3.854 Cylinder1.dat
3 8.3333 13.308 2.7333 4.167 Cylinder2.dat
4 11.7500 13.308 4.1000 4.557 DU40_A17.dat
5 15.8500 11.480 4.1000 4.652 DU35_A17.dat
6 19.9500 10.162 4.1000 4.458 DU35_A17.dat
7 24.0500 9.011 4.1000 4.249 DU30_A17.dat
8 28.1500 7.795 4.1000 4.007 DU25_A17.dat
9 32.2500 6.544 4.1000 3.748 DU25_A17.dat
10 36.3500 5.361 4.1000 3.502 DU21_A17.dat
11 40.4500 4.188 4.1000 3.256 DU21_A17.dat
12 44.5500 3.125 4.1000 3.010 NA64_A17.dat
13 48.6500 2.319 4.1000 2.764 NA64_A17.dat
14 52.7500 1.526 4.1000 2.518 NA64_A17.dat
15 56.1667 0.863 2.7333 2.313 NA64_A17.dat
16 58.9000 0.370 2.7333 2.086 NA64_A17.dat
17 61.6333 0.106 2.7333 1.419 NA64_A17.dat

Integrating the chord distribution along the blade span reveals that the rotor solidity is roughly
5.16%.

As indicated in the table above, the NREL 5MW baseline model incorporates eight unique airfoil
data tables. The two innermost airfoil tables represent cylinders with drag coefficients of 0.50
(Cylinder1.dat) and 0.35 (Cylinder2.dat) and no lift. The remaining six airfoil tables were
created by incorporating 3D corrections to the 2D airfoil data coefficients of the six airfoils used
in the DOWEC study as detailed in Appendix A of Ref. 6 (numerical values for these
coefficients were provided by Koert Lindenburg of ECN). The lift (Cl) and drag (Cd) values
were corrected for rotational stall delay for 0 to 90deg angle-of-attack, the Cd values were
additionally corrected using the Viterna method for 0 to 90deg angle-of-attack, and the Beddoes-
Leishman stall hysteresis parameters were found using version 2.0 of AirfoilPrep3. All
corrections were made assuming an aspect ratio of 17. The 3D corrected airfoil data coefficients
are illustrated graphically in the figures below:

6
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
3D Corrected Coefficients of the DU40 Airfoil
CL
2.5
CD
CM (not req'd.)
2.0

1.5

1.0

0.5

0.0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

-0.5

-1.0

-1.5

3D Corrected Coefficients of the DU35 Airfoil

CL
2.5
CD
CM (not req'd.)
2.0

1.5

1.0

0.5

0.0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

-0.5

-1.0

-1.5

7
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
3D Corrected Coefficients of the DU30 Airfoil
CL
2.5
CD
CM (not req'd.)
2.0

1.5

1.0

0.5

0.0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

-0.5

-1.0

-1.5

3D Corrected Coefficients of the DU25 Airfoil

CL
2.5
CD
CM (not req'd.)
2.0

1.5

1.0

0.5

0.0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

-0.5

-1.0

-1.5

8
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
3D Corrected Coefficients of the DU21 Airfoil
CL
2.5
CD
CM (not req'd.)
2.0

1.5

1.0

0.5

0.0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

-0.5

-1.0

-1.5

3D Corrected Coefficients of the NACA64 Airfoil

CL
2.5
CD
CM (not req'd.)
2.0

1.5

1.0

0.5

0.0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

-0.5

-1.0

-1.5

Nacelle and Hub Properties:

As mentioned earlier, the hub is located 5m upwind of the tower centerline at 90m elevation
above the MSL when the tower is undeflected. The NREL 5MW baseline model uses the same
vertical distance from the tower-top to the hub height used by the DOWEC studythat is, 2.4m

9
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
from Table 6 on page 26 of Ref. 6. The elevation of the yaw bearing is consequently 87.6m.
With a shaft tilt of 5, this makes the distance directed along the shaft from the hub center to the
yaw axis 5.01910m and the vertical distance along the yaw axis from the tower-top to the shaft
1.96256m. The distance directed along the shaft from the hub center to the main bearing is taken
to be 1.912m, from Table 6 on page 26 of Ref. 6.

The hub mass is taken to be 56,780 kg like the REpower 5M and is located at the hub center.
The hub inertia about the shaft is taken to be 115,926 kgm^2. This was found by assuming that
the hub spindel is a thin spherical shell with a radius 0.25m longer than the hub radius (due to the
fact that the pitch bearing is located within the spindel0.25m is chosen since the nacelle height
of the DOWEC 6MW is 3.5m based on Table 6 on page 26 of Ref. 6), which have theoretical
mass moment of inertias of: Ixx = 2/3*m*r^2.

The nacelle mass is taken to be 240,000 kg like the REpower 5M and is located 1.9m downwind
of the yaw axis like the DOWEC 6MW (from Table 7 on page 27 of Ref. 6) and 1.75m above the
yaw bearing, which is half the height of the DOWEC 6MW nacelle (from Table 6 on page 26 of
Ref. 6). The nacelle inertia about the yaw axis is taken to be 2,607.89E3 kgm^2. This was
chosen to be the same as the DOWEC 6MW's nacelle inertia about its nacelle c.g. and translated
to the yaw axis using the parallel axis theorem with the nacelle mass and downwind distance to
the nacelle c.g..

The nacelle yaw actuator is taken to have a natural frequency of 3Hz, which is equivalent to the
highest full structural system natural frequency in the FAST model, and a damping ratio of 2%
critical. This resulted in an equivalent nacelle yaw linear spring constant of 9,028.32E6 N-m/rad
and an equivalent nacelle yaw linear damping constant of 19.16E6 N-m/rad/s.

The nacelle and hub properties discussed in this section are summarized in the table below:

Nacelle and Hub Properties

Elevation of Yaw Bearing Above MSL 87.6 m
Vertical Distance Along Yaw Axis from Yaw Bearing to Shaft 1.96256 m
Distance Along Shaft from Hub Center to Yaw Axis 5.01910 m
Distance Along Shaft from Hub Center to Main Bearing 1.912 m
Hub Mass 56,780 kg
Hub Inertia About Shaft Axis 115,926 kg-m^2
Nacelle Mass 240,000 kg
Nacelle Inertia About Yaw Axis 2,607,890 kg-m^2
Nacelle c.g. Location Downwind of Yaw Axis 1.9 m
Nacelle c.g. Location Above of Yaw Bearing 1.75 m
Equivalent Nacelle Yaw Linear Spring Constant 9,028,320,000 N-m/rad
Equivalent Nacelle Yaw Linear Damping Constant 19,160,000 N-m/rad/s

Drivetrain Properties:
The NREL 5MW baseline wind turbine model uses the same rated rotor speed (12.1 rpm), rated
generator speed (1173.7 rpm), and gearbox ratio (97:1) as the REpower 5M machine. The
gearbox is assumed to have no frictional lossesa requirement of the FAST-to-ADAMS

10
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
preprocessor. The electrical efficiency of the generator is taken to be 94.4%. This is chosen to
be roughly the same as the total mechanical to electrical conversion loss used by the DOWEC
6MW at ratedi.e., 3.5E5 Watts of power loss occur at about 6.25E6 Watts of aerodynamic
power from Figure 15, page 24 of Ref. 6. The generator inertia about the high-speed shaft is
taken to be 534.116 kgm^2, which is the same equivalent low-speed shaft generator inertia used
in the DOWEC studyi.e., 5025500 kgm^2 from page 36 of Ref. 6.

The drive shaft is taken to have the same natural frequency as the RECOFF model and 5%
critical damping ratio for the free-free mode with a rigid generator and rigid rotor. This resulted
in an equivalent drive shaft linear spring constant of 867.637E6 N-m/rad and a damping constant
of 6.215E6 N-m/rad/s.

The high-speed shaft brake is taken to have the same maximum brake torque to maximum
generator torque ratio and the same time lag as used in the DOWEC study (from page 29 of Ref.
6). This resulted in a fully-deployed high-speed shaft brake torque of 28,116.2 N-m and a time
lag of 0.6 seconds. This time lag is the amount of time it takes for the brake to fully engage once
deployed. FAST and ADAMS use a simple linear ramp from 0 to full over the 0.6 seconds.

The drivetrain properties discussed in this section are summarized in the table below:

Drivetrain Properties
Rated Rotor Speed 12.1 rpm
Rated Generator Speed 1173.7 rpm
Gearbox Ratio 97 :1
Electrical Generator Efficiency 94.4 %
Generator Inertia About High-Speed Shaft 534.116 kg-m^2
Equivalent Drive Shaft Torsional Spring Constant 867,637,000 N-m/rad
Equivalent Drive Shaft Torsional Damping Constant 6,215,000 N-m/rad/s
Fully-Deployed High-Speed Shaft Brake Torque 28,116.2 N-m
High-Speed Shaft Brake Time Constant 0.6 sec

Support Structure Properties:

The properties of the support structure depend on whether the turbine will be installed in shallow
water or deep water. The two different support structures are described in separate sections
below.

Shallow Water:
The distributed tower properties of the NREL 5MW baseline model are based on the base
diameter (6m) and thickness (0.027m), top diameter (3.87m) and thickness (0.019m), and
effective mechanical steel properties of the tower used in the DOWEC study, as given in Table 9
on page 31 of Ref. 6. The radius and thickness of the tower are assumed to be linearly tapered
from base to top. The tower is connected to a monopile with a constant diameter of 6m and a
constant thickness of 0.060m. The tower base begins at an elevation of 10m above the MSL.
The monopile extends from here down to the mudline, which is at 20m below MSL. The
Youngs modulus is taken to be 210*10^9 Pa, the shear modulus is taken to be 80.8*10^9 Pa and
the effective density of the steel is taken to be 8500 kg/m^3. The 8500 kg/m^3 is meant to be an

11
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
increase above steels typical value of 7850 kg/m^3 in order to account for paint, bolts, welds,
and flanges that are not accounted for in the thickness data. The resulting distributed tower
properties are given in the table below:

Distributed Tower Properties

Elevation HtFract TMassDen TwFAStif TwSSStif TwGJStif TwEAStif TwFAIner TwSSIner TwFAcgOf TwSScgOf
(m) (-) (kg/m) (Nm^2) (Nm^2) (Nm^2) (N) (kg m) (kg m) (m) (m)
-20.00 0.00000 9517.14 1037.13E+9 1037.13E+9 798.098E+9 235.129E+9 41979.2 41979.2 0.0 0.0
10.00 0.27881 9517.14 1037.13E+9 1037.13E+9 798.098E+9 235.129E+9 41979.2 41979.2 0.0 0.0
10.00 0.27882 4306.51 474.49E+9 474.49E+9 365.133E+9 106.396E+9 19205.6 19205.6 0.0 0.0
17.76 0.35094 4030.44 413.08E+9 413.08E+9 317.878E+9 99.576E+9 16720.0 16720.0 0.0 0.0
25.52 0.42306 3763.45 357.83E+9 357.83E+9 275.356E+9 92.979E+9 14483.4 14483.4 0.0 0.0
33.28 0.49517 3505.52 308.30E+9 308.30E+9 237.242E+9 86.607E+9 12478.7 12478.7 0.0 0.0
41.04 0.56729 3256.66 264.08E+9 264.08E+9 203.220E+9 80.459E+9 10689.2 10689.2 0.0 0.0
48.80 0.63941 3016.86 224.80E+9 224.80E+9 172.987E+9 74.534E+9 9098.9 9098.9 0.0 0.0
56.56 0.71153 2786.13 190.06E+9 190.06E+9 146.252E+9 68.834E+9 7692.7 7692.7 0.0 0.0
64.32 0.78365 2564.46 159.49E+9 159.49E+9 122.735E+9 63.357E+9 6455.7 6455.7 0.0 0.0
72.08 0.85576 2351.87 132.77E+9 132.77E+9 102.167E+9 58.105E+9 5373.9 5373.9 0.0 0.0
79.84 0.92788 2148.34 109.54E+9 109.54E+9 84.291E+9 53.077E+9 4433.6 4433.6 0.0 0.0
87.60 1.00000 1953.87 89.49E+9 89.49E+9 68.863E+9 48.272E+9 3622.1 3622.1 0.0 0.0

The entries in the first column, Elevation, are the vertical locations along the tower centerline
relative to the MSL. HtFract is the fractional height along the tower centerline from the mudline
(0.0) to the tower top (1.0). The rest of columns are similar to those described for the distributed
blade properties.

The resulting overall (integrated) tower + monopile mass is 522,617 kg and is centered at
37.172m along the tower centerline above the mudline. This result follows directly from the
overall tower height of 87.6m.

The NREL 5MW baseline model incorporates 1% critical structural damping in all modes of the
isolated tower (without the top mass present), which corresponds to the values used in the
DOWEC study from page 21 of Ref. 6.

The table below summarizes the undistributed tower properties discussed in this section:

Undistributed Tower Properties

Tower-Top Height Above MSL 87.6 m
Tower-Base Height Above MSL 10 m
Water Depth (From MSL) 20 m
Overall (Integrated) Mass 522,617 kg
c.g. Location (w.r.t. Mudline Along Tower Centerline) 37.172 m
Structural Damping Ratio (All Modes) 1 %

Deep Water:
The distributed tower properties of the NREL 5MW baseline model are based on the base
diameter (6m) and thickness (0.027m), top diameter (3.87m) and thickness (0.019m), and
effective mechanical steel properties of the tower used in the DOWEC study, as given in Table 9
on page 31 of Ref. 6. The Youngs modulus is taken to be 210*10^9 Pa, the shear modulus is
taken to be 80.8*10^9 Pa and the effective density of the steel is taken to be 8500 kg/m^3. The
8500 kg/m^3 is meant to be an increase above steels typical value of 7850 kg/m^3 in order to
account for paint, bolts, welds, and flanges that are not accounted for in the tower thickness data.

12
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
The radius and thickness of the tower are assumed to be linearly tapered from base to top. The
thickness of the tower was scaled up relative to the values indicated above in order to strengthen
the tower since the REpower 5M machine has a larger tower-top mass than the DOWEC 6MW
machine. An increase of 30% was chosen to ensure that the first tower fore-aft (FA) and side-to-
side (SS) tower frequencies lie between 1P and 3P through the operational range of the wind
turbine in a Campbell diagram. The resulting distributed tower properties are given in the table
below:

Distributed Tower Properties

Elevation HtFract TMassDen TwFAStif TwSSStif TwGJStif TwEAStif TwFAIner TwSSIner TwFAcgOf TwSScgOf
(m) (-) (kg/m) (Nm^2) (Nm^2) (Nm^2) (N) (kg m) (kg m) (m) (m)
0.00 0.0 5590.87 614.34E+9 614.34E+9 472.75E+9 138.13E+9 24866.3 24866.3 0.0 0.0
8.76 0.1 5232.43 534.82E+9 534.82E+9 411.56E+9 129.27E+9 21647.5 21647.5 0.0 0.0
17.52 0.2 4885.76 463.27E+9 463.27E+9 356.50E+9 120.71E+9 18751.3 18751.3 0.0 0.0
26.28 0.3 4550.87 399.13E+9 399.13E+9 307.14E+9 112.43E+9 16155.3 16155.3 0.0 0.0
35.04 0.4 4227.75 341.88E+9 341.88E+9 263.09E+9 104.45E+9 13838.1 13838.1 0.0 0.0
43.80 0.5 3916.41 291.01E+9 291.01E+9 223.94E+9 96.76E+9 11779.0 11779.0 0.0 0.0
52.56 0.6 3616.83 246.03E+9 246.03E+9 189.32E+9 89.36E+9 9958.2 9958.2 0.0 0.0
61.32 0.7 3329.03 206.46E+9 206.46E+9 158.87E+9 82.25E+9 8356.6 8356.6 0.0 0.0
70.08 0.8 3053.01 171.85E+9 171.85E+9 132.24E+9 75.43E+9 6955.9 6955.9 0.0 0.0
78.84 0.9 2788.75 141.78E+9 141.78E+9 109.10E+9 68.90E+9 5738.6 5738.6 0.0 0.0
87.60 1.0 2536.27 115.82E+9 115.82E+9 89.13E+9 62.66E+9 4688.0 4688.0 0.0 0.0

The entries in the first column, Elevation, are the vertical locations along the tower centerline
relative to the ground or MSL. HtFract is the fractional height along the tower centerline from
the tower base (0.0) to the tower top (1.0). The rest of columns are similar to those described for
the distributed blade properties.

The resulting overall (integrated) tower mass is 347,460 kg and is centered at 38.234m along the
tower centerline above the MSL. This result follows directly from the overall tower height of
87.6m.

The NREL 5MW baseline model incorporates 1% critical structural damping in all modes of the
isolated tower (without the top mass present), which corresponds to the values used in the
DOWEC study from page 21 of Ref. 6.

The table below summarizes the undistributed tower properties discussed in this section:

Undistributed Tower Properties

Height Above MSL 87.6 m
Overall (Integrated) Mass 347,460 kg
c.g. Location (w.r.t. MSL Along Tower Centerline) 38.234 m
Structural Damping Ratio (All Modes) 1 %

Overall Body Mass Matrix of Rigid Turbine:

In the preliminary stages of offshore platform design, it is convenient to use frequency-domain
analysis techniques. Such techniques generally consider only the six rigid body modes of the
platform (surge, sway, heave, roll, pitch, and yaw) and ignore the flexibility of the system
(platform and wind turbine). Knowledge of the 6x6 body mass matrix of the 5MW baseline

13
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
wind turbine is required for such analyses. As defined in equation 141 on page 149 of Ref. 15,
the body mass matrix of the rigid, undeflected 5MW baseline wind turbine in deep water to four-
significant digits is:

697,500 0 0 0 44,630,000 0
0 697,500 0 44,630,000 0 144,800

0 0 697,500 0 144,800 0
Mij =
0 44,630,000 0 3,566,000,000 0 11,670,000
44,630,000 0 144,800 0 3,551,000,000 0

0 144,800 0 11,670,000 0 25,410,000

This matrix was derived by linearizing the FAST model about a reference point located at the
tower centerline at the base of the tower (i.e., MSL). The row/column order is: 1 = surge, 2 =
sway, 3 = heave, 4 = roll, 5 = pitch, 6 = yaw and the units in the matrix are: kg, kg-m, and kg-
m^2.

Naturally, since FAST does not have a tower torsion DOF, neither does it have knowledge of the
yaw inertia of the tower about the tower centerline. So, the effects of the tower yaw inertia are
not included in the body mass matrix given above. By integration of the distributed tower inertia
data along the length of the tower, this yaw inertia is estimated to be 2,241,820 kg-m^2. This
inertia could be added to element (6,6) of the body mass matrix given above, but it is not
straightforward how this additional inertia will affect the other elements in the matrix. However,
neglecting the tower yaw inertia from the body mass matrix should not be of significant concern,
since it may be seen that the addition of the tower yaw inertia to element (6,6) will only increase
the overall turbine yaw inertia (tower, nacelle, and rotor) by a small amount (less than 10%).
Users of the body mass matrix should be aware of this issue and address it as they see fit.

Variable-Speed Generator Torque and Blade Pitch Control Properties:

The NREL 5MW baseline wind turbine incorporates a control system that consists of a variable-
speed generator torque controller and a collective blade pitch controller. Both controllers use the
generator speed measurement as the sole input. The goal of the controller is to maximize power
capture below, and regulate speed above, the rated operating point. The control system is
implemented as an external dynamic-link-library (DLL) in the style of Garrad Hassans
BLADED wind turbine software package. The control system does not attempt to regulate the
nacelle yaw angle nor does it attempt to control the deployment of the high speed shaft brake. A
flowchart of the control system calculations is given in the following figure.

14
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
Control System Flowchart

The generator speed measurement for both the torque and pitch controllers is filtered using a
recursive, single-pole, low-pass filter with exponential smoothing. The corner frequency (-3dB
point) of the low pass filter is set to 0.25 Hz, which is chosen to be roughly of the blade
edgewise natural frequency.

The generator torque is computed as a function of the filtered generator speed using a variable-
speed controller. The variable-speed controller incorporates five control regions: 1, 1, 2, 2,
and 3. Region 1 is the control region before cut-in where the generator torque is zero so that no
power is extracted from the wind; instead, the wind is used to accelerate the rotor. Region 2 is
the control region for optimizing power capture. Here, the generator torque is proportional to the
square of the filtered generator speed. In region 3, the generator power is held constant so that
the generator torque is inversely proportional to the filtered generator speed. Region 1 is the
start-up region and is a linear transition between regions 1 and 2. It is used to place a lower limit
on the generator speed in order to limit the operational speed range. Region 2 is a linear
transition between regions 2 and 3 with a torque slope corresponding to the slope of an induction
machine. Region 2 is typically needed (as is the case for this 5MW model) since a machine
does not typically reach rated torque at its rated speed using Region 2s control law [i.e., the
optimal K is typically lower than that which would make the rated torque = K*(rated speed)^2,
since the rated speed is generally limited from optimal in order to limit tip speed for noise
reasons].

15
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006

The peak of the power coefficient as a function of tip-speed ratio and blade pitch surface was
found by running FAST simulations at a number of fixed rotor speeds and a number of fixed
rotor-collective blade pitch angles at a fixed wind speed of 8 m/s. From these simulations, the
peak power coefficient of 0.482 was found to occur at a tip-speed ratio of 7.55 and a rotor-
collective blade pitch of 0.0. With the 97:1 gearbox ratio, this resulted in an optimal constant of
proportionality of 0.0255764 N-m/rpm^2 in the region 2 control law. With the rated generator
speed of 1173.7 rpm, rated electric power of 5MW, and a generator efficiency of 94.4%, the
rated mechanical power is 5.296610MW and the rated generator torque is 43,093.55 N-m.
Region 1 spans the range of generator speeds between 670 rpm and 30% above this value, or
871 rpm. The minimum generator speed of 670 rpm corresponds to the minimum rotor speed of
6.9 rpm used by the REpower 5M machine16. The transitional generator speed between Regions
2 and 3 is taken to be 99% of the rated generator speed, or 1,161.963 rpm. The generator slip
percentage in region 2 is taken to be 10% in accordance with the value used in the DOWEC
studyreference page 24 of Ref. 6. The resulting generator torque versus generator speed curve
is shown in the figure below:

Variable-Speed Control System Torque versus Speed Curve

50000

45000
Optimal
40000
Variable Speed Controller

35000
Generator Torque, Nm

30000

25000

20000

15000

10000

5000

0
0 200 400 600 800 1000 1200 1400
Generator Speed, rpm

Finally, a conditional statement is placed on the variable speed controller so that the torque is
computed as if it was in Region 3 regardless of the generator speed whenever the previous pitch
angle command is greater or equal to 1. This results in improved output power quality (fewer
dips below rated) at the expense of short term overloading of the generator and gearbox. To
avoid excessive overloading of both, the torque is saturated to a maximum of 10% above rated,
or 47,402.91 N-m. Also, a torque rate limit of 15,000 N-m/s is imposed.

16
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
The collective blade pitch angle commands are computed using gain-scheduled PI control on the
speed error between the filtered generator speed and the rated generator speed (1173.7 rpm).
The parameter for determining the gain scheduling correction factor is the previous pitch angle
command. The gains (see the figure below) were derived according to the technique
documented in Ref. 4. In particular, the partial derivative of the rotor power with respect to the
collective pitch angle as a function of the pitch angle was computed by performing a
linearization analysis in FAST at a number of fixed pitch angles. The frozen wake assumption
was used when computing these partial derivatives through a minor modification to AeroDyn.

Pitch Control System Gain-Scheduling Law

1.0 0.025

0.9 Gain Correction Factor

Proportional Gain
0.8 Integral Gain 0.020

0.7
Correction Factor

0.6 0.015

Gains
0.5

0.4 0.010

0.3

0.2 0.005

0.1

0.0 0.000
0 5 10 15 20
Collective Blade Pitch Angle (deg)

The pitch rate limit is set at 8/sec in absolute value. This is speculated to be the pitch rate limit
of conventional 5MW machines based on GE Winds long blade test program. The minimum
and maximum pitch settings are chosen to be 0 and 90, respectively. The integral term in the
PI controller is saturated between these limits so as to ensure fast response between the Region 2
/ Region 3 transition.

Due to limitations in the code, the FAST model does not include any pitch actuator dynamic
effects. However, pitch actuator dynamics are required in ADAMS. Hence, in the ADAMS
model, the blade pitch actuator is taken to have a very high natural frequency of 30Hz, which is
equivalent to 10x the highest full structural system natural frequency in the FAST model, and a
damping ratio of 2% critical. This resulted in an equivalent blade pitch linear spring constant of
1,069.02E6 N-m/rad and an equivalent blade pitch linear damping constant of 0.23E6 N-m/rad/s.

17
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006

The variable-speed generator-torque and blade pitch control properties discussed in this section
are summarized in the table below:

Variable-Speed Generator-Torque and Blade Pitch Control Properties

Corner Frequency of Generator Speed Low Pass Filter 0.25 Hz
Peak Power Coefficient 0.482
Tip Speed Ratio at Peak Power Coefficient 7.55
Rotor Collective Blade Pitch Angle at Peak Power Coefficient 0.0
Generator Torque Constant in Region 2 0.0255764 N-m/rpm^2
Rated Mechanical Power 5.296610 MW
Rated Generator Torque 43,093.55 N-m
Transitional Generator Speed Between Regions 1 and 1 1/2 670 rpm
Transitional Generator Speed Between Regions 1 1/2 and 2 871 rpm
Transitional Generator Speed Between Regions 2 1/2 and 3 1,161.963 rpm
Generator Slip Percentage in Region 2 1/2 10 %
Minimum Pitch For Ensuring Region 3 Torque 1
Maximum Generator Torque 47,402.91 N-m
Maximum Generator Torque Rate 15,000 N-m/s
Proportional Gain at Minimum Pitch Setting 0.01882681 sec
Integral Gain at Minimum Pitch Setting 0.008068634
Pitch Angle at Which Rotor Power Has Doubled 6.302336
Minimum Pitch Setting 0
Maximum Pitch Setting 90
Maximum Absolute Pitch Rate 8 /sec
Equivalent Blade Pitch Linear Spring Constant 1,069,020,000 N-m/rad
Equivalent Blade Pitch Linear Damping Constant 230,000 N-m/rad/s

Degrees-of-Freedom and Timesteps:

Not including platform motion, the FAST model incorporates 16 degrees-of-freedom (DOFs) as
follows:
2 flap and 1 edge mode DOFs for each of the 3 blades
1 variable-speed generator DOF and 1 driveshaft torsional DOF
1 nacelle yaw actuator DOF
2 fore-aft and 2 side-to-side mode DOFs in the tower

The ADAMS model incorporates 378 DOFs as follows:

102 DOFs, including flap and edge shear and bending, torsion, and extension DOFs, in
each of the 3 blades
1 pitch actuator DOF in each of the 3 blades
1 variable-speed generator DOF and 1 driveshaft torsional DOF
1 nacelle yaw actuator DOF
66 DOFs, including fore-aft and side-to-side shear and bending, torsion, and extension
DOFs, in the tower

Time integration occurs at a constant time step of 0.0125 seconds in FASTs fixed-step size
integration scheme and at a maximum step size of 0.0125 seconds in ADAMS variable-step size

18
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
integration scheme. Aerodynamic calculations are performed every other time step (0.025
seconds) to ensure that there are at least 200 azimuth step computations per revolution. These
time steps are the largest possible to ensure numerical stability across a range of operating
modes.

Full-System Natural Frequencies:

In order to provide a cursory view of the structural response of the full system, NREL performed
an eigenanalysis on the stationary turbine (not spinning) in the deepwater configuration in order
to obtain full system natural frequencies. The eigenanalysis ignores gravitational and
aerodynamic loads, as well as structural damping. Results for the first 13 full system natural
frequencies are listed in the table below. In FAST, these are obtained by performing an
eigenanalysis in MATLAB on the first order state matrix created from a FAST linearization
analysis. In ADAMS, the frequencies are obtained by invoking a LINEAR/EIGENSOL
command, which linearizes the complete ADAMS model and computes eigendata. The ADAMS
results were obtained without mass offsets in the blade in order to ensure consistency between
the FAST and ADAMS models. The agreement between FAST and ADAMS is quite good.

Full System Natural Frequencies at 0 RPM, Hz

Mode Description FAST ADAMS
1 1st Tower SS 0.31780 0.31620
2 1st Tower FA 0.31994 0.31930
3 1st Drivetrain 0.62149 0.61100
4 1st Blade Asym. Flapwise Yaw 0.67250 0.63840
5 1st Blade Asym. Flapwise Pitch 0.67084 0.66900
6 1st Blade Collective Flap 0.70130 0.70230
7 1st Blade Asym. Edgewise Pitch 1.07770 1.07510
8 1st Blade Asym. Edgewise Yaw 1.09270 1.08860
9 2nd Blade Asym. Flapwise Yaw 1.99320 1.70440
10 2nd Blade Asym. Flapwise Pitch 1.93830 1.85760
11 2nd Blade Collective Flap 2.01980 1.96180
12 2nd Tower SS 2.94270 2.93910
13 2nd Tower FA 2.91650 2.84620

Wind Speed Relationships and Initial Conditions:

The figures below contain many basic machine specifications as a function of wind speed, such
as rotor/generator speed, torque, thrust, pitch, etc. These were obtained by running a number of
FAST simulations at many fixed wind speeds. The simulations were run using the EQUILibrium
wake option of AeroDyn with all DOFs enabled. These relationships can be used, for example,
to set initial values of rotor speed and blade pitch in simulations.

19
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
Wind Speed Relationships Legend
BldPitch1 Pitch Angle of Blade 1
GenPwr Electrical Generator Power
GenSpeed Angular Speed of the High-Speed Shaft and Generator
GenTq Electrical Generator Torque
IPDefl1 In-Plane Tip Deflection of Blade 1
OoPDefl1 Out-of-Plane Tip Deflection of Blade 1
RotPwr Mechanical Rotor Power
RotSpeed Angular Speed of the Low-Speed Shaft and Rotor
RotThrust Rotor Thrust
RotTorq Mechanical Rotor Torque
TSR Rotor Blade Tip Speed Ratio
TTDspFA Fore-Aft Displacement of the Tower Top and Yaw Bearing
TTDspSS Side-to-Side Displacement of the Tower Top and Yaw Bearing

6000
Region 1 2 2 3

GenSpeed (rpm)
RotPwr (kW)
5000
GenPwr (kW)
RotThrust (kN)
RotTorq (kNm)
4000

3000

2000

1000

0
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind Speed (m/s)

20
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
50
Region 1 2 2 3

45 RotSpeed (rpm)
BldPitch1 (deg)
40 GenTq (kNm)
TSR (-)
35

30

25

20

15

10

0
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind Speed (m/s)

6
Region 1 2 2 3
OoPDefl1 (m)
5
IPDefl1 (m)
TTDspFA (m)
4 TTDspSS (m)

-1

-2
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind Speed (m/s)

References:
1
De Vries, Eize, Multibrid: 'A New Offshore Wind Turbine Contender',
http://www.jxj.com/magsandj/rew/2004_05/multibird.html, Renewable Energy World, Vol. 7,
No. 5, James & James (Science Publishers) Ltd, London, United Kingdom, September-October
2004.
2
Goezinne, F., Terms of reference DOWEC, 176-FG-R0300, DOWEC 10041_000, September
2001.

21
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
3
Hansen, Craig, NWTC Design Codes: AirfoilPrep,
http://wind.nrel.gov/designcodes/preprocessors/airfoilprep/, Golden, CO: National Renewable
Energy Laboratory, October 2004.
4
Hansen, Morten H.; Hansen, Anca; Larsen, Torben J.; ye, Stig; Srensen; and Fuglsang, Peter,
Control Design for a Pitch-Regulated, Variable-Speed Wind Turbine, Ris-R-1500(EN),
Roskilde, Denmark: Ris National Laboratory, January, 2005.
5
Jonkman, J.M. and Buhl, M.L., Jr., FAST User's Guide, NREL/EL-500-29798, Golden, CO:
National Renewable Energy Laboratory, October 2004.
6
Kooijman, H.J.T., Lindenburg, C., Winkelaar, D., and van der Hooft, E.L., DOWEC 6 MW
Pre-Design: Aero-elastic modeling of the DOWEC 6 MW pre-design in PHATAS, ECN-CX--
01-135, DOWEC 10046_009, Petten, the Netherlands: Energy Research Center of the
Netherlands, September 2003.
7
Laino, David J. and Hansen, A. Craig, User's Guide to the Computer Software Routines
AeroDyn Interface for ADAMS, Salt Lake City, UT: Windward Engineering LLC, Prepared
for the National Renewable Energy Laboratory under Subcontract No. TCX-9-29209-01,
September 2001.
8
Lindenburg, C., Aeroelastic Modelling of the LMH64-5 Blade, DOWEC-02-KL-083/0,
DOWEC 10083_001, Petten, the Netherlands: Energy Research Center of the Netherlands,
December 2002.
9
LM Glasfiber Group, Wind Turbine Blades, Product Overview, Standard Products Max.
Rated Power <=5000 kW,
http://www.lmglasfiber.dk/UK/Products/Wings/ProductOverView/50000kw.htm, Lunderskov,
Denmark: LM Glasfiber Group, January 2005.
10
Malcolm, D.J. and Hansen, A.C., WindPACT Turbine Rotor Design Study, NREL/SR-500-
32495, Golden, CO: National Renewable Energy Laboratory, August 2002.
11
Multibrid Technology, Technical Data Multibrid M5000,
http://www.multibrid.com/download/Datenblatt_M5000_eng.pdf, Bremerhaven, Germany:
Multibrid Technology, March 2004.
12
Multibrid Technology, The Concept in Detail, http://www.multibrid.com/english/concept.htm,
Bremerhaven, Germany: Multibrid Technology, January 2005.
13
Musial, Walt, Butterfield, Sandy, and Boone, Andrew, Feasibility for Floating Platform
Systems for Wind Turbines, Collection of the 2004 ASME Wind Energy Symposium Technical
Papers Presented at the 42nd AIAA Aerospace Sciences Meeting and Exhibit, 58 January 2004,
Reno, NV, Washington, D.C.: American Institute of Aeronautics and Astronautics, January 2004,
pp. 476486; NREL/CP-500-36504.

22
 NRELOffshrBsline5MW.doc
Jason Jonkman, NREL/NWTC
June 14, 2006
14
National Renewable Energy Laboratory, About the Program: WindPACT,
http://www.nrel.gov/wind/windpact/, Golden, CO: National Renewable Energy Laboratory,
January 2005.
15
Newman, J.N., Marine Hydrodynamics, Cambridge, MA: The Massachusetts Institute of
Technology Press, 1977.
16
REpower Systems, REpower 5M,
http://www.repower.de/typo3/fileadmin/download/produkte/5m_uk.pdf, Hamburg, Germany:
REpower Systems, January 2005.
17
REpower Systems, REpower Systems AG Renewable Energy for the Future,
http://www.repower.de/, Hamburg, Germany: REpower Systems, January 2005.
18
Smith, Kevin, WindPACT Turbine Design Scaling Studies; Technical Area 2: Turbine, Rotor,
and Blade Logistics, NREL/SR-500-29439, Golden, CO: National Renewable Energy
Laboratory, June 2001.
19
Tarp-Johansen, Neils Jacob, RECOFF Home Page, http://www.risoe.dk/vea/recoff/, Roskilde,
Denmark: Ris National Laboratory, July 2004.

23