March 2020: IEA Wind TCP Task 37

March 2020
IEA Wind TCP Task 37
Definition of the IEA Wind

15-Megawatt Offshore
Reference Wind Turbine
Technical Report
A iea wind
Definition of the IEA 15-Megawatt
Offshore Reference Wind
Evan Gaertner1, Jennifer Rinker2, Latha Sethuraman1,
Frederik Zahle2, Benjamin Anderson1, Garrett Barter1,
Nikhar Abbas1, Fanzhong Meng2, Pietro Bortolotti1,
Witold Skrzypinski2, George Scott1, Roland Feil1,
Henrik Bredmose2, Katherine Dykes2, Matt Shields1,
Christopher Allen3, and Anthony Viselli3
1 National Renewable Energy Laboratory
2 Technical University of Denmark
3 University of Maine
Suggested Citation
Gaertner, Evan, Jennifer Rinker, Latha Sethuraman, Frederik Zahle, Benjamin Anderson,
Garrett Barter, Nikhar Abbas, Fanzhong Meng, Pietro Bortolotti, Witold Skrzypinski,
George Scott, Roland Feil, Henrik Bredmose, Katherine Dykes, Matt Shields, Christopher
Allen, and Anthony Viselli. 2020. Definition of the IEA 15-Megawatt Offshore Reference
Wind. Golden, CO: National Renewable Energy Laboratory. NREL/TP-5000-75698.
https://www.nrel.gov/docs/fy20osti/75698.pdf
NREL is a national laboratory of the U.S. Department of Energy Technical Report

Office of Energy Efficiency & Renewable Energy NREL/TP-5000-75698
Operated by the Alliance for Sustainable Energy, LLC March 2020
This report is available at no cost from the National Renewable Energy National Renewable Energy Laboratory
Laboratory (NREL) at www.nrel.gov/publications. 15013 Denver West Parkway
Golden, CO 80401
Contract No. DE-AC36-08GO28308 303-275-3000 • www.nrel.gov
NOTICE
This work was authored in part by the National Renewable Energy Laboratory, operated by Alliance for Sustainable
Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding
provided by U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Wind and Water
Technologies Office. The views expressed herein do not necessarily represent the views of the DOE or the U.S.
Government.
This report is available at no cost from the National Renewable

Energy Laboratory (NREL) at www.nrel.gov/publications.
U.S. Department of Energy (DOE) reports produced after 1991

and a growing number of pre-1991 documents are available
free via www.OSTI.gov.
NREL prints on paper that contains recycled content.

Acknowledgements
Staff from the National Renewable Energy Laboratory (NREL) and the Technical University of Denmark (DTU) col-
laborated closely, via the International Energy Agency (IEA) Wind Task 37 on Systems Engineering in Wind Energy,
to design the reference turbine described in this report. The authors from the University of Maine (UMaine) lever-
aged their expertise and resources to contribute the floating platform semisubmersible design. To better capture the
individual contributions, and as a resource for future inquiries, the following table lists the roles and responsibilities
for all of the authors involved.
Name Institution Contribution
Evan Gaertner NREL Primary design engineer who led the blade, tower, and monopile design
Jennifer Rinker DTU HAWC2 lead, design load basis, controller tuning
Latha Sethuraman NREL Designer of permanent-magnet direct-drive generator, drivetrain, bedplate,
nacelle, and other subsystems
Frederik Zahle DTU Bend-twist coupling contribution, rotor and blade design review
Benjamin Anderson NREL Created nacelle CAD model and performed drivetrain and bedplate analysis
Garrett Barter NREL Project principal investigator
Nikhar Abbas NREL Reference OpenSource Controller lead and tuning
Fanzhong Meng DTU Controller lead and tuning
Pietro Bortolotti NREL Blade design support
Witold Skrzypiński DTU Tool development for blade design
George Scott NREL Drivetrain design support
Roland Feil NREL Detailed blade structural analysis
Henrik Bredmose DTU COREWIND principal investigator
Katherine Dykes DTU General support
Matt Shields NREL Monopile and transition piece design support, report editing
Christopher Allen UMaine Lead semisubmersible design engineer
Anthony Viselli UMaine Semisubmersible principal investigator
Beyond IEA Wind facilitated collaboration between NREL, DTU, and UMaine, the larger networks of individual
staff members and the IEA Wind Task 37 effort were leveraged to ensure that the design represented a conservative
estimate of industry capabilities. These industry contacts gave invaluable information to calibrate our design as-
sumptions and input values. Without their input, the 15-MW reference turbine would not be nearly as professional
of a design or as useful to the broader community. In no particular order, we extend our thanks to the following
companies and individuals:
• General Electric: Anna Diedrichkeit, Albert Fiass, Vanita Mani, Priyan Subramanian, David Torrey, Yogen
Utturkar
• EDF Renewables: Julien Simon
• Senvion: Fabian Vorpahl
• Sintef: Karl Merz
• Atkins: Ikpoto Udoh
• NREL: Walt Musial, Jason Jonkman, Nicole Mendoza
• Sandia National Laboratories: Ernesto Camarena.
The work at NREL was sponsored by the U.S. Department of Energy Office of Energy Efficiency and Renewable
Energy Wind Energy Technologies Office under the guidance of program manager, Patrick Gilman. The research at
DTU received funding from the European Union’s H2020 Program under Grant Agreement no. 815083 - COREWIND.
i
This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications
Nomenclature
Acronyms
3D three-dimensional
BECAS BEam Cross section Analysis Software
DLC design load case
DTU Technical University of Denmark
HAWTOpt2 Horizontal Axis Wind Turbine Optimization 2nd generation
HAWC2 Horizontal Axis Wind turbine simulation Code 2nd generation
IEC International Electrotechnical Commission
IEA International Energy Agency
metocean meteorological ocean
NREL National Renewable Energy Laboratory
NdFeB neodymium
PSD power spectral density
PI proportional integral
ROSCO Reference OpenSource Controller
RMS root mean squared
SRB spherical roller bearing
SST shear stress transport
TDO tapered double outer
TSR tip-speed ratio
UMaine University of Maine
WindPACT Wind Partnership for Advanced Component Technology
WISDEM® Wind-Plant Integrated System Design Engineering Model
WP work package
Units
A ampere
h hour
Hz hertz
kg kilogram
m meter
min minute
N (kgm/s2 ) Newton
rad radian
rpm revolutions per minute
P period
Pa (N/m2 ) pascal
s second
t metric tonne
T tesla
V volt
W watt
Prefixes
m milli
k kilo
M mega
G giga
ii
Executive Summary
Overview and Motivation

Reference wind turbines serve multiple roles within the wind community and have therefore grown in importance
in recent years. First, they serve as open benchmarks that are defined with publicly available design parameters to
be used as baselines for studies that explore new technologies or design methodologies. Second, as an open design,
reference wind turbines enable collaboration between industry and external researchers. Finally, reference wind
turbines offer an entry point and educational platform for newcomers to wind energy to understand fundamental
design elements and system trade-offs.
For fixed-bottom offshore wind energy, the average turbine size for European deployment in 2018 was 6.8 MW [1],
and GE will launch its 12-MW Haliade-X offshore turbine to the market in 2021 with a rotor diameter of 218 m
and direct-drive configuration. To be relevant now and in the coming years, a new reference wind turbine should
leap ahead of the current generation of industry wind turbines, but cannot leap so far that aggressive technology
innovations are required. Therefore, a reference wind turbine above 10 MW [2], yet below 20 MW [3], is needed,
that continues on the same growth trend as the GE Haliade-X using a similar drivetrain configuration and specific
power.
This report describes a 15-megawatt (MW) offshore wind turbine with a fixed-bottom monopile support structure.
This reference wind turbine is a Class IB direct-drive machine, with a rotor diameter of 240 meters (m) and a hub
height of 150 m. An overview of the design is presented in Figure ES-1 and Table ES-1. The design reflects a joint
effort between the National Renewable Energy Laboratory (NREL), sponsored by the U.S. Department of Energy,
and the Technical University of Denmark (DTU), sponsored by the European Union’s H2020 Program, through the
second work package of International Energy Agency (IEA) Wind Task 37 on Wind Energy Systems Engineering:
Integrated RD&D. A forthcoming report will detail a semisubmersible floating support structure developed by the
University of Maine (UMaine).
Blade and Rotor Properties

Top level rotor configuration decisions were informed by discussions with industry partners, on what would be tech-
nically feasible for the next generation of wind turbines. The blade design was driven by the selection of 240 m as
the rotor diameter and a maximum tip speed of 95 meters per second (m/s). A fairly traditional structural configu-
ration was selected, comprising of two main load-carrying, carbon-reinforced spars, connected by two shear webs,
with reinforcement along the trailing and leading edge and foam fillers. The DTU FFA-W3 series of airfoils were
used due to their publicly available polars and geometries. The blade chord, twist, airfoil positions, tip speed ratio,
and spar cap thickness were selected through a design optimization study. Table ES-2 summarized key features of
the blades, including a design power coefficient, CP , of 0.489 and 65 metric tons (t) of blade mass.
Tower and Monopile Properties

The tower and monopile were designed as an isotropic steel tube. Frequency considerations constrained much of the
design in that the first tower-monopile mode, 0.17 hertz (Hz), lies between the 1P and 3P blade passing frequencies
for all wind speeds. This is also sufficient to avoid the range of highest energy ocean wave frequencies for a generic
East Coast site (0.10 Hz to 0.13 Hz). The tower height was chosen such that the hub height reaches 150 m, allowing
for 30 m of ground (water surface) clearance with up to 120-m blades. The monopile foundation has a 10-m outer
diameter, which pushes the limits of current manufacturing and installation technology, and a thickness profile that
varies from 55 millimeters (mm) in the pile to 44 mm at the transition piece.
Nacelle and Drivetrain Properties

The 15-MW reference wind turbine uses a direct-drive layout with a permanent-magnet, synchronous, radial flux
outer-rotor generator in a simple and compact nacelle layout. Figure ES-2a shows a simple direct-drive nacelle
layout with an outer-rotor permanent-magnet generator. The assembly consists of a hub shaft supporting the turbine
iii
Table ES-1. Key Parameters for the IEA Wind 15-MW Turbine
Parameter Units Value
Power rating MW 15
Turbine class - IEC Class 1B
Specific rating W/m2 332
Rotor orientation - Upwind
Number of blades - 3
Control - Variable speed

Collective pitch
Cut-in wind speed m/s 3
Rated wind speed m/s 10.59
Cut-out wind speed m/s 25
Design tip-speed ratio - 90
Minimum rotor speed rpm 5.0
Maximum rotor speed rpm 7.56
Maximum tip speed m/s 95
Rotor diameter m 240

Airfoil series - FFA-W3
Hub height m 150
Hub diameter m 7.94
Hub overhang m 11.35
Rotor precone angle deg -4.0
Blade prebend m 4
Blade mass t 65
Drivetrain - Direct drive

Shaft tilt angle deg 6
Rotor nacelle assembly mass t 1,017
Transition piece height m 15

Monopile embedment depth m 45
Monopile base diameter m 10
Tower mass t 860
Monopile mass t 1,318
deg degrees rpm revolutions per minute
m meters t metric tons
m/s meters per second W/m2 watts per square meter
iv
Figure ES-1. The IEA Wind 15-MW reference wind turbine
and generator rotors on two main bearings housed on a stationary turret that is cantilevered from the bedplate. The
hub is a simple spherical shell, with cutouts for the blades and the flange. The main shaft has a hollow cylindrical
cross section, with a constant wall thickness and a tilt angle of 6◦ . The main shaft, along with the rotor, is supported
by two main bearings. Both these main bearings have rotating outer raceways and fixed inner raceways. The outer
raceways and bearing housing are accommodated by a turret held by the bedplate. The entire weight of the turbine
rotor, generator rotor, and hub loads are transmitted by the main shaft to the turret via the bearings. The bedplate is
a hollow, elliptically curved, cantilever beam with circular cross sections. The yaw system bearings are double-row,
angular, contact ball bearings.
The generator construction features an external rotor radial flux topology machine with a surface-mounted permanent
magnet (shown in Figure ES-2b). The outer rotor layout facilitates a simple and rugged structure, easy manufac-
turing, short end windings, and better heat transfer between windings and teeth than the inner rotor configuration.
The stator design features fractional, slot-layout, double-layer concentrated coils, which maximize the fundamental
winding factor.
Load Analysis
This work assumes a generic U.S. East Coast site with a wind speed described by a Weibull distribution with a mean
velocity of approximately 8.65 m/s and a shape parameter of 2.12. At this mean wind speed, the corresponding
significant wave height is approximately 1.4 m, with a peak spectral period of 7.9 seconds (s). The fixed-bottom
v
Table ES-2. Blade Properties
Description Value Units
Blade length 117 m

Root diameter 5.20 m
Root cylinder length 2.34 m
Max chord 5.77 m
Max chord spanwise position 27.2 m
Tip prebend 4.00 m
Precone 4.00 deg
Blade mass 65,250 kg
Blade center of mass 26.8 m
Design tip-speed ratio 9.00 -
First flapwise natural frequency 0.555 Hz
First edgewise natural frequency 0.642 Hz
Design CP 0.489 -
Design CT 0.799 -
Annual energy production 77.4 GWh
deg degrees kg kilograms
GWh gigawatt-hours m meters
Hz Hertz
(a)
(b)
Figure ES-2. A sketch and CAD model of the nacelle layout of the 15-MW direct-drive wind turbine. Not to scale
and some structural details omitted. Blades (not shown), hub, shaft, and generator rotor rotate.
vi
monopile support presented in this report is designed around a water depth of 30 m.
An International Electrotechnical Commission design load case [4] analysis study was conducted to determine
the worst-case ultimate loading on key design constraining components. Yaw-misaligned parked conditions with
extreme wind speeds and extreme coherent gust with a direction change result in the worst-case loading for this
design. The worst-case out-of-plane tip deflection is 22.8 m, leaving more than sufficient tower clearance, with
an unbent blade tip-to-tower clearance of 30.0 m. This margin suggests that the blade design is conservative and
further aeroelastic optimization could potentially improve the aerodynamic performance or cost of energy while still
remaining within recommended safety margins. A full fatigue analysis of this blade was not conducted, which could
potentially be an issue for the edgewise blade bending moments for very large blades.
Availability
To foster further collaboration, the reference turbine design is available for use by the broader wind energy com-
munity in input files that support a variety of analysis tools, including OpenFAST, HAWC2, the Wind-Plant In-
tegrated System Design & Engineering Model (WISDEM), and HawtOpt2. These files are hosted on GitHub at
github.com/IEAWindTask37/IEA-15-240-RWT, with the intent that the community will contribute back to the effort
by submitting their design variants for inclusion in the repository.
vii
Contents
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1 The Role of Reference Wind Turbines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Overall Turbine Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Design Tools and Methodologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 Model Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5 Meteorological Ocean Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Blade Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1 Blade Aerodynamic Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Blade Structural Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3 Rotor Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1 Controller Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2 Steady-State Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4 Tower and Monopile Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5 Nacelle, Drivetrain, and Hub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

5.1 Nacelle Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.2 Hub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3 Main Shaft, Bearings, and Turret . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.4 Bedplate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.5 Yaw System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.6 Direct-Drive Generator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.7 Nacelle Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6 Load Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
6.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
A Blade Cross Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
B Tower and Soil Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

B.1 Tower Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
B.2 Soil Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
List of Figures
Figure ES-1.The IEA Wind 15-MW reference wind turbine . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
1
Figure ES-2.A sketch and CAD model of the nacelle layout of the 15-MW direct-drive wind turbine. Not to
scale and some structural details omitted. Blades (not shown), hub, shaft, and generator rotor rotate. . . . vi
Figure 1-1. The IEA Wind 15-MW reference wind turbine . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 2-1. View from the suction side (top) and trailing edge (bottom) of the offshore wind turbine blade . . 8
Figure 2-2. DTU FFA-W3 airfoil family used in the IEA Wind 15-MW blade design . . . . . . . . . . . . . . 9
Figure 2-3. Aerodynamic polars for the airfoils used on the blade . . . . . . . . . . . . . . . . . . . . . . . . 10
Figure 2-4. Blade planform spanwise quantities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Figure 2-5. Lofted blade shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Figure 2-6. Blade cross section at 70% span . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Figure 2-7. Schematic of IEA Wind Turbine Ontology composite definition, from root to tip . . . . . . . . . . 13
Figure 2-8. Blade layup layer thickness as a function of the normalized s-coordinate around the airfoil at
various span positions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 2-9. Shear layup layer thickness at various span positions . . . . . . . . . . . . . . . . . . . . . . . . 15
Figure 2-10. Blade beam structural properties versus the blade-curve position along the span computed using
PreComp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Figure 2-11. Blade planform and structural properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Figure 3-1. OpenFAST blade element momentum performance and operation of the 15-MW rotor with the
ROSCO controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Figure 3-2. CCBlade steady-state blade element momentum aerodynamic power and thrust coefficient sur-
faces as a function of blade pitch and TSR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Figure 3-3. Spanwise forces on the blade as a function of wind speed . . . . . . . . . . . . . . . . . . . . . . 20
Figure 4-1. Tower natural frequency relative to the normalized power spectral density (PSD) of the excitation
frequencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Figure 4-2. Outer diameter and wall thickness for tower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Figure 4-3. Tower and monopile cross-sectional properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Figure 5-1. A sketch and CAD model of the nacelle layout of the 15-MW direct-drive wind turbine. Not to
scale and some structural details omitted. Blades (not shown), hub, shaft, and generator rotor rotate. . . . 25
Figure 5-2. CAD illustration of (a) the main shaft and (b) turret (also called the nose); dimensions are docu-
mented in Table 5-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Figure 5-3. A CAD illustration of the bedplate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Figure 5-4. A CAD illustration of an outer rotor direct-drive generator with electromagnetic and structural
design parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Figure 6-1. DLC ranking of maximum blade root and tower base bending moments . . . . . . . . . . . . . . 33
Figure 6-2. DLC ranking of maximum blade tip and tower top deflections . . . . . . . . . . . . . . . . . . . 33
Figure A-1. Blade cross section at 0% span . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Figure A-2. Blade cross section at 10% span . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2
Figure A-10.Blade cross section at 100% span . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
List of Tables
Table ES-1.Key Parameters for the IEA Wind 15-MW Turbine . . . . . . . . . . . . . . . . . . . . . . . . . . iv

Table ES-2.Blade Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Table 1-1. Key Parameters for the IEA Wind 15-MW Turbine, As Compared to the DTU 10-MW Turbine . . . 5
Table 1-2. Models Used for Design and Analysis of the IEA Wind 15-MW Reference Wind Turbine . . . . . . 7
Table 2-1. Blade Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Table 4-1. Material Properties for the Tower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Table 4-2. Some Key Properties and Dimensions of the Tower and Foundation . . . . . . . . . . . . . . . . . 23
Table 5-1. Lumped Masses and Moments of Inertia for the Nacelle Assembly . . . . . . . . . . . . . . . . . 26
Table 5-2. Main Shaft Dimensions, Bearing, and Loads Used in Sizing . . . . . . . . . . . . . . . . . . . . . 27
Table 5-3. Bedplate Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Table 5-4. Electromagnetic and Structural Design of the 15-MW Direct-Drive Generator . . . . . . . . . . . . 31
Table 6-1. Summary of IEC DLC Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Table 6-2. Metocean Conditions Used in DLC Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3
1 Introduction
1.1 The Role of Reference Wind Turbines

Reference wind turbines serve multiple roles within the wind community and have therefore grown in importance in
recent years. First, they serve as open benchmarks that are defined with publicly available design parameters to be
used as baselines for studies that explore new technologies or design methodologies. Traditionally, reference wind
turbines have been realistic, but not fully optimized, designs so that they can be updated and improved upon by the
active wind energy community. Second, as an open design, reference wind turbines enable collaboration between
industry and external researchers. By using a reference turbine, industry can protect its intellectual property yet
still explore advanced technology development with outsiders. Finally, reference wind turbines offer an entry point
and educational platform for newcomers to wind energy to understand fundamental design elements and system
trade-offs.
The history of reference wind turbines begins in the early 2000s with the National Renewable Energy Laboratory
(NREL) Wind Partnership for Advanced Component Technology (WindPACT) turbine series, which includes 0.75-,
1.5-, and 3-megawatt (MW) turbines [5]. Their use, however, was restricted to national laboratories in the United
States. The first widely adopted reference turbine by the larger international community was the NREL 5-MW tur-
bine [6], which is still used by many researchers today. More recently, the Technical University of Denmark (DTU)
developed a 10-MW turbine for offshore wind applications [7]. These two turbines have been supplemented by other
turbines, such as an 8-MW turbine in the European Union FP7 project LEANWIND [8], the Sandia National Labo-
ratories’ 100-meter (m)-blade studies [9], and a conceptual study of a 20-MW turbine in the INNWIND project [3].
Most recently, the IEA Wind Task 37, which coordinated this effort, also released modernized 3.35-MW land-based
and 10-MW offshore reference turbines [2]. These designs have been released quickly on the heels of one another
as the industry has rapidly increased the power rating and size of its product lines. For fixed-bottom offshore wind
energy, the average turbine size for European deployment in 2018 was 6.8 MW [1], and GE will launch its 12-MW
Haliade-X offshore turbine to the market in 2021 with a rotor diameter of 218 m and direct-drive configuration.
To be relevant now and in the coming years, a new reference wind turbine must leap ahead of the current generation
of industry wind turbines, but cannot leap so far that aggressive technology innovations are required. The current
slate of reference wind turbine designs cannot fully meet the needs of the research community and industry to ad-
vance the state of the art in blade scaling, floating foundation design, wind farm control, logistic studies, and many
other topics. Therefore, a reference wind turbine above 10 MW, yet below 20 MW, is needed that continues on the
same growth trend as the GE Haliade-X using a similar drivetrain configuration and specific power.
This is the motivation for the design effort of this IEA Wind 15-MW reference wind turbine described in this report.
This reference wind turbine, Figure 1-1, is a Class IB direct-drive machine, with a rotor diameter of 240 m and a hub
height of 150 m. The design reflects a joint effort between NREL, sponsored by the U.S. Department of Energy, and
DTU, sponsored by the European Union’s H2020 Program, through the second work package of IEA Wind Task 37
on Wind Energy Systems Engineering: Integrated RD&D. This report describes an offshore fixed-bottom monopile
support structure, with a forthcoming report to detail a semisubmersible floating support structure developed in
collaboration with the University of Maine (UMaine).
1.2 Overall Turbine Parameters

The overall parameters for the turbine are stated in Table 1-1. The table also shows the data for the DTU 10-MW
reference wind turbine [7] for comparison.
1.3 Design Tools and Methodologies

The IEA Wind 15-MW reference turbine was jointly designed by NREL, DTU, and UMaine. The analysis and de-
sign tools that were leveraged as part of this effort are listed in Table 1-2. These model names will appear frequently
in the discussion of the design in the sections to come.
4
Table 1-1. Key Parameters for the IEA Wind 15-MW Turbine, As Compared to the DTU 10-MW Turbine
Parameter Units DTU 10-MW Turbine IEA Wind 15-MW Turbine
Power rating MW 10 15
Turbine class - IEC Class 1B IEC Class 1B
Specific rating W/m2 401 332
Rotor orientation - Upwind Upwind
Number of blades - 3 3
Control - Variable speed Variable speed
- Collective pitch Collective pitch
Cut-in wind speed m/s 4 3
Rated wind speed m/s 11.4 10.59
Cut-out wind speed m/s 25 25
Rotor diameter m 178.3 240

Airfoil series - FFA-W3 FFA-W3
Hub height m 119 150
Hub diameter m 5.6 7.94
Hub overhang m 7.1 11.35
Drivetrain - Medium speed Low speed

- Multiple-stage gearbox Direct drive
Design tip-speed ratio - 75 90
Minimum rotor speed rpm 6.0 5.0
Maximum rotor speed rpm 9.6 7.56
Maximum tip speed m/s 90 95
Gearbox ratio - 50 —
Shaft tilt angle deg 5 6

Rotor precone angle deg -2.5 -4.0
Blade prebend m 3.332 4
Blade mass t 41 65
Rotor nacelle assembly mass t 674 1,017
Tower mass t 987 860
Tower base diameter m 8 10
Transition piece height m 10 15

Monopile embedment depth m 42.6 45
Monopile base diameter m 9 10
Monopile mass t 2,044 1,318
deg degrees rpm revolutions per minute
m meters t metric tons
m/s meters per second W/m2 watts per square meter
MW megawatts
5
Figure 1-1. The IEA Wind 15-MW reference wind turbine
6
Table 1-2. Models Used for Design and Analysis of the IEA Wind 15-MW Reference Wind Turbine
Role NREL Tool Chain DTU Tool Chain
System Design WISDEM [10], [11] HAWTOpt2 [12], [13]

CCBlade [14]
RotorSE [15]
DrivetrainSE
TowerSE
Preprocessors PreComp [16] BECAS [17]
BModes [18]
Aeroelastic Analysis OpenFAST [19], [20] HAWC2 [21]
HAWCStab2 [22]
Most of the design was conducted within the Wind-Plant Integrated System Design & Engineering Model (WISDEM® ),
which is a family of models that are generally simplified and quasi-static to enable rapid design optimization at a
limited number of design points. WISDEM is built on top of National Atmospheric and Space Administrations’s
OpenMDAO library, which drives the optimization and serves as the glue code between different models [23]. Con-
ceptual designs were verified and enriched with more complete load and performance analysis using the nonlinear
transient models of OpenFAST, HAWC2, and HAWCStab2. The results of these higher-fidelity simulations were
used to update the design variable bounds and constraint values within WISDEM, and the process was iterated.
1.4 Model Availability

The reference turbine design is available for use by the broader wind energy community in input files that support a
variety of analysis tools, including OpenFAST, HAWC2, WISDEM, and HawtOpt2. Additionally, the data depicted
in graphs and tables in this report are also available electronically, in Microsoft Excel format, instead of writing them
out as appendices. These files are hosted on GitHub at:
• github.com/IEAWindTask37/IEA-15-240-RWT
• github.com/IEAWindTask37/IEA-15-240-RWT/blob/master/Documentation.
The open-source availability of the IEA Wind 15-MW reference wind turbine is intended to encourage the commu-
nity to contribute back to the effort by submitting their design variants for inclusion into the repository and further
use by others.
1.5 Meteorological Ocean Environment

As a generic reference turbine, the design is intended to apply to many different offshore locations. However, the
analysis of ultimate loads and the design of the substructure depend on the particular wind, wave, and soil profiles.
The work of Stewart et al. [24] provides a general yet specific enough meteorological ocean (metocean) environ-
ment to execute the analysis and design. This work assumes a generic U.S. East Coast site, with detailed wind and
wave probability distributions found in the repository documentation listing described in the previous subsection. As
a quick summary, the wind speed is described as a Weibull distribution with parameters [9.767, 2.12], which gives a
mean velocity of approximately 8.65 meters per second (m/s). At this mean wind speed, the corresponding signifi-
cant wave height is approximately 1.4 m, with a peak spectral period of 7.9 seconds (s). The fixed-bottom monopile
support presented in this report is designed around a water depth of 30 m.
7
2 Blade Properties
The blade length of this IEA Wind 15-MW reference turbine is 117 m with a root diameter of 5.2 m and a maximum
chord of 5.77 m at approximately 20% span. The overall blade mass is around 65 metric tons (t) and is designed to
achieve a power coefficient, CP , of 0.489. A top-down and edge view of the blade are shown in Figure 2-1 and a
more complete statistical breakdown is listed in Table 2-1.
Figure 2-1. View from the suction side (top) and trailing edge (bottom) of the offshore wind turbine blade
Table 2-1. Blade Properties
Description Value Units
Blade length 117 m

Root diameter 5.20 m
Root cylinder length 2.34 m
Max chord 5.77 m
Max chord spanwise position 27.2 m
Tip prebend 4.00 m
Precone 4.00 deg
Blade mass 65,250 kg
Blade center of mass 26.8 m
Design tip-speed ratio 9.00 -
First flapwise natural frequency 0.555 Hz
First edgewise natural frequency 0.642 Hz
Design CP 0.489 -
Design CT 0.799 -
Annual energy production 77.4 GWh
deg degrees kg kilograms
GWh gigawatt-hours m meters
Hz Hertz
2.1 Blade Aerodynamic Properties

The DTU FFA-W3 series of airfoils for use in the blade design. These are publicly available and well-documented
airfoils that were also used in the IEA Wind/DTU 10-MW offshore reference wind turbine and are shown in Figure
2-2.
The airfoil data for each of the FFA-W3 airfoils was generated at a Reynolds number of Re = 107 . To compute the
aerodynamic coefficients in the range of −32° to 32°, we used the two-dimensional incompressible Navier-Stokes
solver, EllipSys2D [25–27]. The meshes were generated using HypGrid2D [28], with a 512-by-256-cell radial
grid. Simulations assumed fully turbulent and freely transitioning boundary layers, based on the k − ω shear stress
transport (SST) turbulence model [29] and the Drela-Giles transition model [30], assuming a freestream turbulence
intensity of 0.1%. We performed a 360° extrapolation using AirfoilPreppy, but three-dimensional (3D) corrections
were not applied to the polars because the spanwise distribution of relative thickness was a free design variable. A
Du-Selig [31] stall delay 3D correction was applied to the polar data for the OpenFAST model of the final design.
Figure 2-3 shows the aerodynamic characteristics of the airfoils used on the blade. Tabular data of airfoil shapes and
performance polars are provided in the parallel spreadsheet documentation.
8
0.2
0.1
0.0
0.1
0.2
0.0 0.2 0.4 0.6 0.8 1.0
SNL-FFA-W3-500 FFA-W3-241 FFA-W3-301 FFA-W3-360
FFA-W3-211 FFA-W3-270blend FFA-W3-330blend
Figure 2-2. DTU FFA-W3 airfoil family used in the IEA Wind 15-MW blade design
The blade planform design variables are plotted in Figure 2-4. The aerodynamic center of the airfoils is used for
the blade pitch axis. There are a number of aerodynamic design characteristics that are worth noting. The transition
from a cylinder cross section to the thickest 50% airfoil occurs between 2.34 m to 17.55 m or 2%–15% of the span,
with the maximum chord of 5.77 m at 27.2 m of span (23.3%). This is shown in the chord and relative thickness
profiles in Figure 2-4a–b. With such a large blade radius, the design was heavily driven by the tip deflection loading
and tower clearance constraint. The twist profile in Figure 2-4c shows some unloading at the blade tip, which sheds
some energy production to mitigate the strongest thrust loads at the most flexible part of the blade. This behavior
will be evident again in the rotor performance plots in Section 3. The blade was designed with a significant prebend
away from the tower to provide additional tip clearance, with 4 m separating the tip chordline from the root (Figure
2-4d). When axially stacking the airfoils to generate the lofted blade shape, the cant angle from prebend curvature is
not considered. More prebend would have given further margin, reducing stiffness requirements, but the value was
limited to 4 m based on blade molding and other manufacturing challenges, as communicated by industry. Advanced
manufacturing techniques may enable greater blade prebend in the future, but this is a reasonable constraint at this
time.
2.2 Blade Structural Properties

The lofted blade shape is shown in Figure 2-5 and an internal structural layout at 70% span is shown in Figure 2-6,
with additional spanwise locations in Appendix A. The structural layout of the blade is fairly traditional, comprising
two main load-carrying spars placed on a straight line connecting the root and the tip, along with reinforcement
along the trailing and leading edges. One of these spar caps is placed on the airfoil pressure side and the other on
the suction side. These spar caps are made out of carbon fiber to provide as much stiffness with as little weight
as possible. The blade has two shear webs that connect the pressure side and suction side, attached to the main
spars, extending from a 10% to 95% span; shown in Figure 2-6 as the vertical members. Leading and trailing edge
reinforcements are also added using uniaxial glass fiber to provide additional edgewise stiffness. Foam filler panels
were added between the leading-edge and trailing-edge reinforcement and the spar caps, on both the pressure side
and suction side.
9
2.0
1.5 SNL-FFA-W3-500 Re=8.1e6
FFA-W3-211 Re=1e7
Lift coefficient, cl
1.0 FFA-W3-241 Re=1e7
FFA-W3-270blend Re=1e7
0.5 FFA-W3-301 Re=1e7
0.0 FFA-W3-360 Re=1e7
0.5
1.0
20 15 10 5 0 5 10 15 20
Angle of Attack [deg]
(a) Airfoil lift coefficients
2.0
1.5 SNL-FFA-W3-500 Re=8.1e6
FFA-W3-211 Re=1e7
Lift coefficient, cl
1.0 FFA-W3-241 Re=1e7

0.5 FFA-W3-301 Re=1e7
0.0 FFA-W3-360 Re=1e7
0.5
1.0
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
Drag coefficient, cd
(b) Airfoil lift-drag polars
125
100 SNL-FFA-W3-500 Re=8.1e6
Lift over drag, cl/cd
FFA-W3-211 Re=1e7
75 FFA-W3-241 Re=1e7
50 FFA-W3-301 Re=1e7
25 FFA-W3-330blend Re=1e7
FFA-W3-360 Re=1e7
0
25
50
20 15 10 5 0 5 10 15 20
Angle of Attack [deg]
(c) Airfoil lift-to-drag coefficients
Figure 2-3. Aerodynamic polars for the airfoils used on the blade
10
6
100
Relative Thickness [%]

5 90
80
4
Chord [m]
70
3 60
50
2 40
1 30
20
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Nondimensional Blade Span (r/R) Nondimensional Blade Span (r/R)
(b) Relative thickness
(a) Chord length
15.0 0
12.5
1
10.0 Prebend [m]
Twist [deg]
7.5
2
5.0
2.5 3
0.0
2.5 4
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
(c) Twist (d) Prebend
Pitch Axis Chord Location [%]
50
45
40
35
30
0.0 0.2 0.4 0.6 0.8 1.0
Nondimensional Blade Span (r/R)
(e) Chordwise offset
Figure 2-4. Blade planform spanwise quantities
11
Figure 2-5. Lofted blade shape
Figure 2-6. Blade cross section at 70% span
12
The internal structure and composite layup of the blade is defined according to the IEA Wind Turbine Ontology [32].
Composite layers are defined as spanwise elements superimposed on the blade shell or shear webs, following the
curved blade reference axis. The wind turbine ontology allows for multiple methods of defining elements, dimen-
sionally, using the layer width (arc length), offset, and rotation relative to a reference position, or nondimensionally,
using the normalized arc length positions, as shown in Figure 2-7. The normalized arc length position coordinate (s)
is defined as zero at the suction-side trailing edge and as one at the pressure-side trailing edge. For flatback airfoils,
the trailing edge is defined as the midpoint of the flatback surface.
Line
Arc Width s = 0 Chord
Offset s=1
rot Plane of Rotation
Pitch Axis
Figure 2-7. Schematic of IEA Wind Turbine Ontology composite definition, from root to tip
The material layup of the blade is plotted in Figures 2-8 and 2-9, shown along the airfoil shell as a function of the arc
length s-coordinate and for the shear webs. The complete layup definition of the structural components is provided
in the accompanying blade ontology and Microsoft Excel files. With the composite layup defined, the blade beam
structural properties were computed with PreComp [16] and VABS [33, 34] (in the NREL tool chain) or BECAS [17,
35] (in the DTU tool chain). Specifically, these tools calculated the stiffness matrices for each cross section along
the blade, which were then used in OpenFAST or HAWC2. A comparison of the turbine performance between the
NREL and DTU modeling tools is discussed in Rinker et al. [36]. Figures 2-10 and 2-11 show the resulting blade
beam structural properties. The structural damping of the first flapwise, edgewise, and torsional modes were assumed
to be 3%, 3%, and 6%, respectively.
13
120 CarbonUD glass_biax glass_uni FFA-W3-360 airfoil, 25% span
Gelcoat glass_triax medium_density_foam
Thickness [mm] 100
80
60
40
20
0
TE Suction Side LE Pressure Side TE
Airfoil s-coordinate [-]
CarbonUD glass_biax glass_uni FFA-W3-301 airfoil, 44% span
120 Gelcoat glass_triax medium_density_foam
Thickness [mm]
100
80
60
40
20
0
80 FFA-W3-211 airfoil, 77% span
CarbonUD glass_biax glass_uni
60
Thickness [mm]
50
40
30
20
10
0
CarbonUD glass_biax glass_uni FFA-W3-211 airfoil, 100% span
Thickness [mm]
25
20
15
10
5
0
Figure 2-8. Blade layup layer thickness as a function of the normalized s-coordinate around the airfoil at various span positions
14
Thickness [mm]
0
10
20
30
40
50
60
circular 0%
circular 2%
SNL-FFA-W3-500 15%
Gelcoat
CarbonUD
FFA-W3-360 25%
FFA-W3-330blend 33%
web0
FFA-W3-301 44%
FFA-W3-270blend 54%
FFA-W3-241 64%
glass_biax
glass_triax
FFA-W3-211 77%
15
FFA-W3-211 100%
circular 0%
circular 2%
glass_uni
SNL-FFA-W3-500 15%
FFA-W3-360 25%
FFA-W3-330blend 33%
FFA-W3-301 44%
medium_density_foam
web1
Figure 2-9. Shear layup layer thickness at various span positions
FFA-W3-270blend 54%
FFA-W3-241 64%
FFA-W3-211 77%
FFA-W3-211 100%
1.6 1e11
3000 Edge
1.4 Flap
Mass density [kg/m]
2500 Torsional
Stiffness [N.m^2]
1.2
2000 1.0
1500 0.8
1000 0.6
0.4
500
0.2
0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
(a) Mass density
(b) Edgewise, flapwise, and torsional stiffness
1e10
4
Axial Stiffness [N]
0
0.0 0.2 0.4 0.6 0.8 1.0
Nondimensional Blade Span (r/R)
(c) Axial stiffness
Figure 2-10. Blade beam structural properties versus the blade-curve position along the span computed using PreComp
2
1
Chordwise [m]
0
1
2 Pitch axis
Mass center
3 Neutral center
Geometric center
Shear center
4
0.0 0.2 0.4 0.6 0.8 1.0
Blade span r/R
Figure 2-11. Blade planform and structural properties
16
3 Rotor Performance
Rotor performance is tightly coupled with the controller behavior and pitch schedule to reach the desired shaft
revolutions per minute (rpm), torque, and thrust. The controller is described before presenting rotor performance
data.
3.1 Controller Properties

Two controllers are provided for the IEA Wind 15-MW reference wind turbine: the NREL Reference OpenSource
Controller (ROSCO) [37] and the DTU Basic Controller [38], which are implemented in OpenFAST and HAWC2,
respectively. The rotor operates with a minimum rotational speed of 5 rpm to avoid 3-period (3P) interference with
the tower/monopile natural frequency, and reaches a rated rotational speed of 7.55 rpm at 10.59 m/s, resulting in
a maximum nominal tip speed of 95 m/s. The rotor operates with a pitch setting of 0° at the design tip-speed ratio
(TSR), but operates with positive pitch at low wind speeds to track maximum power while maintaining the minimum
rotor speed. The rotor starts pitching at the rated wind speed of 10.59 m/s.
3.1.1 Controller Methodology
In both controllers, two active proportional integral (PI) controllers are implemented for the generator torque and
blade pitch angles. Saturation limits on rotor speeds and blade pitch angles are implemented to ensure turbine opera-
tion within the design constraints. The controller operation can be distinguished by three regions:
3 m/s ≤ U ≤ 6.98 m/s; minimum rotor speed. A PI controller on the generator torque is used to regulate the tur-
bine to the turbine’s minimum rotor speed, 5 rpm. In ROSCO, the minimum blade pitch angle is defined based
on a wind speed estimate, such that CP is maximized. The CP -maximizing minimum blade pitch angles are
found a-priori using steady-state blade element momentum analysis, provided by CCBlade [14]. For the DTU
Basic Controller, a PI controller is applied to the torque controller to regulate the rated wind speed. The pitch
angles are determined using a look-up table scheduled on the filtered hub-height wind speed.
6.98 m/s ≤ U ≤ 10.59 m/s; optimal TSR. In below-rated wind speeds, the rotor speed is regulated to operate at the
turbine’s optimal TSR with a PI controller on the generator torque.
10.59 m/s ≤ U ≤ 25 m/s; rated power In above-rated wind speeds, the rotor speed is regulated via a PI controller
on the blade pitch angle. The objective of the blade pitch controller is to regulate the rotor speed to its rated
value, 7.55 rpm. For floating applications, the constant-power setting is traditionally replaced by constant
generator torque.
In ROSCO, an extended Kalman-filter-based wind speed estimator is employed to estimate the optimal rotor speeds
in below-rated operation and to define minimum blade pitch angles during minimum rotor speed operation. In
Region 2.5, instead of the linear constant speed approach used by the NREL 5-MW reference turbine, the ROSCO
control uses a setpoint-smoothing methodology [39] to ensure smooth transitions between control regions. As a
result of the design optimization process, the IEA Wind 15-MW turbine has a negligible Region 2.5, since the rated
rotor speed is reached at rated torque. The PI gains for the generator torque and blade pitch controllers are calculated
using the ROSCO generic tuning methodology [39] and included in the accompanying OpenFAST input files.
The parameters for the DTU Basic Controller were determined using HAWCStab2’s controller tuning feature. The
assumed natural frequencies and damping for the partial-load poles (torque controller) and full-load poles (pitch
controller) were 0.05 Hz, 70% critical, and 0.03 Hz, 70% critical, respectively. Quadratic gain scheduling was used
with an assumption of constant power. The resulting controller parameters are found in the HAWC2 input files
and the definitions for all parameters and Regions 1, 2, and 3 are consistent with the definitions in the DTU Basic
Controller report [38].
17
3.2 Steady-State Performance
Figure 3-1 shows the steady-state performance of the rotor as a function of wind speed, using OpenFAST with the
ROSCO controller. In Region 1.5, minimum rotor speed constraints result in significantly higher, suboptimal tip-
speed ratios. The blade pitch controller imposes a minimum pitch constraint based on the wind speed estimate, so
that CP is maximized. This results in positive blade pitch angles of up to 4° at low wind speeds. Concurrently, the
generator torque controller attempts to set the rotor to the minimum rotor speed in Region 1.5. In Region 2, the
torque controller tracks the set point tip-speed ratio, which is set near or at maximum CP . The design point for this
blade is T SR =9.0 and blade pitch Θ pitch =0°, which is slightly below the optimal of T SR =8.5 and Θ pitch =−1.0°,
shown in the CP and CT curves in Figure 3-2. This slightly suboptimal design point is a result of the blade design
process, as a trade-off between power production and design constraining loads. The set point smoothing routine
in ROSCO prevents contradictions between the generator torque and blade pitch controllers during Region 2.5
transitions. The Region 2.5 for this design is effectively negligible, so there is little influence from the set point
smoothing and peak shaving routines in ROSCO. Finally, in Region 3, the torque controller is saturated at rated
torque and the blade pitch PI controller pitches to feather to maintain rated rotor speed.
Further analysis of the blade performance is shown in Figure 3-3. Here, the distributed aerodynamic forces on the
blade are computed over a range of below-rated wind speeds. In Section 2, the aerodynamic design of the blade,
especially the twist, was presented as tightly constrained by the tip deflection and tower clearance constraints. Forces
on the blade peak at approximately 90 m and then begin to decrease toward the tip. This unloading of the blade tip
helps prevent tower strikes; however, it results in lost power production, especially because the blade tip has the
highest aerodynamic efficiency and marginal swept area. The sharp, linear drop for the outboard 5% of the blade is
caused by the Prandtl tip loss model employed in the aeroelastic simulations.
18
Region Reg. Region
1.5 2 3
20
10
0
3 5 10 15 20 25
Wind Speed [m/s]
Blade Pitch [ ] Rotor Speed [RPM] Gen. Torque [MN-m] TSR[-]
(a) Controller regulation trajectory
Region Reg. Region

1.5 2 3
20 3.00
Generator Power [MW]
Rotor Thrust [MN]

15 2.25
10 1.50
5 0.75
0 0.00
3 5 10 15 20 25
Wind Speed [m/s]
Generator Power [MW] Rotor Thrust [MN]
(b) Power and thrust curve
Region Reg. Region

1.5 2 3
1.0
Design Point
CP = 0.489
CT = 0.804
0.5
0.0
3 5 10 15 20 25
Wind Speed [m/s]
Power Coef. [-] Thrust Coef [-]
(c) Aerodynamic performance coefficients
Figure 3-1. OpenFAST blade element momentum performance and operation of the 15-MW rotor with the ROSCO controller
19
Power Coefficient Thrust Coefficient
11 .40
0 11
0
50
0.600
0
00
Tip Speed Ratio, [-]
Tip Speed Ratio, [-]

60
0.75
0
1.0
0.380
.38 0.4
0.9
10 0 40 10
0.4
0.450
9 0.420 9
0.400
70
0.4 80
8 0.450 0.4 8
0 2
40
0.4
7 0.4 7
5.0 2.5 0.0 2.5 5.0 5.0 2.5 0.0 2.5 5.0
Blade Pitch Angle, [ ] Blade Pitch Angle, [ ]
Design Point
Figure 3-2. CCBlade steady-state blade element momentum aerodynamic power

and thrust coefficient surfaces as a function of blade pitch and TSR
Tangential Force [N/m]
10000 800
Normal Force [N/m]
7500 600
5000 400
2500 200
0 0
0.0 0.5 1.0 0.0 0.5 1.0
Blade Span [m] Blade Span [m]
U = 6 m/s U = 7 m/s U = 8 m/s U = 9 m/s U = 10 m/s
Figure 3-3. Spanwise forces on the blade as a function of wind speed
20
4 Tower and Monopile Properties
The tower and monopile were designed as an isotropic steel tube in WISDEM and OpenFAST. The monopile design
was informed by the helpful guide published by Arany [40]. Frequency considerations constrained much of the de-
sign in that the first tower-monopile mode, 0.170 Hz, lies between the 1P and 3P blade passing frequency ranges for
all wind speeds, as shown in Figure 4-1. This is also sufficient to avoid the range of expected ocean wave frequencies
for the generic East Coast site: 0.10 Hz to 0.13 Hz. Note that when a floating substructure is used, the frequency
requirements and boundary conditions will shift such that a stiffer tower is required. Therefore, the tower geometry
presented here will be replaced with one better suited for a semisubmersible platform.
1.00
Normalized PSD
0.75
0.50
0.25
0.00
0.0 0.1 0.2 0.3 0.4 0.5
Frequency [Hz]
Wind, Waves, Tower, 1P 3P
Kaimal Spect. JONSWAP Spect. 1st Nat. Freq.
Figure 4-1. Tower natural frequency relative to the normalized power spectral density (PSD) of the excitation frequencies
The tower height was chosen such that the hub height reaches 150 m, allowing for 30 m of ground (water surface)
clearance with 120-m blades.1 The monopile foundation has a 10-m outer diameter, which pushes the limits of
current manufacturing and installation technology. The thickness and outer diameter are shown in Figure 4-2 and
tabulated in Table 4-2, down through the embedded suction pile. The material properties are stated in Table 4-1.
A detailed geotechnical analysis of soil properties was not conducted as part of this design effort. Instead, the soil
foundation was modeled with a series of spring constants to represent the soil stiffness in a one-dimensional finite-
element model. The selected summary values for soil shear modules and Poisson’s ratio are representative of dense
sand or gravel soils [41]. The equations for the soil stiffness sprint constants are found in Appendix B.2.
Table 4-1. Material Properties for the Tower
Parameter Symbol Value Units
Young’s modulus, steel E 2 00E11 Pa

Shear modulus, steel G 7 93E10 Pa
Density, steel ρ 7 85E3 kg/m3
Shear modulus, soil Gs 140E6 Pa
Poisson’s ratio, soil νs 0.4
Pa pascal kg/m3 kilogram per cubic meter
Both HAWC2 and OpenFAST require a number of structural cross-sectional properties that can be easily determined
from the diameter and wall thickness profiles. These properties are shown in Figure 4-3, with the underlying equa-
tions provided in Appendix B. Note that the final mass values reported in Table 1-1 include an extra 7% of outfitting
mass beyond the cylindrical shell mass and 100 t for the transition piece.
1 The 30-m clearance value is not specified in an official standard and offshore wind turbine clearances can vary between 20 m to 30 m.
21
150
100
Tower Height [m]
50
Tower transition
0 Water line
Mud line
50
6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 20 25 30 35 40 45 50 55

Outer Diameter [m] Wall Thickness [mm]
Figure 4-2. Outer diameter and wall thickness for tower
14000 Fore-aft/side-side inertia [kg.m]

160000
Mass Density [kg/m]
12000
140000
10000 120000
8000 100000
80000
6000 60000
4000 40000
20000
50 0 50 100 150 50 0 50 100 150
Location [m] Location [m]
(a) Mass density (b) Fore-aft and side-side inertia
1e12 1e11
Fore-aft/side-side 3.5
4.0 Torsional
3.0
Axial Stiffness [N]
3.5
Stiffness [N.m^2]
3.0 2.5
2.5
2.0 2.0
1.5 1.5
1.0
0.5 1.0
50 0 50 100 150 50 0 50 100 150
Location [m] Location [m]
(c) Flap, edgewise, and torsional stiffness (d) Axial stiffness
Figure 4-3. Tower and monopile cross-sectional properties
22
Table 4-2. Some Key Properties and Dimensions of the Tower and Foundation
Location Height [m] Outer Diameter [m] Thickness [mm]
Monopile start -75.000 10.000 55.341

Mud line -30.000 10.000 55.341
-29.999 10.000 55.341
-25.000 10.000 55.341
-24.999 10.000 53.449
-20.000 10.000 53.449
-19.999 10.000 51.509
-15.000 10.000 51.509
-14.999 10.000 49.527
-10.000 10.000 49.527
-9.999 10.000 47.517
-5.000 10.000 47.517
-4.999 10.000 45.517
Water line 0.000 10.000 45.517
0.001 10.000 43.527
5.000 10.000 43.527
5.001 10.000 42.242
10.000 10.000 42.242
10.001 10.000 41.058
Tower start 15.000 10.000 41.058
15.001 10.000 39.496
28.000 10.000 39.496
28.001 10.000 36.456
41.000 9.926 36.456
41.001 9.926 33.779
54.000 9.443 33.779
54.001 9.443 32.192
67.000 8.833 32.192
67.001 8.833 30.708
80.000 8.151 30.708
80.001 8.151 29.101
93.000 7.390 29.101
93.001 7.390 27.213
106.000 6.909 27.213
106.001 6.909 24.009
119.000 6.748 24.009
119.001 6.748 20.826
132.000 6.572 20.826
132.001 6.572 23.998
Tower top 144.582 6.500 23.998
23
The damping for the tower is stiffness-proportional only. For both OpenFAST and HAWC2, the stiffness-proportional
factors were determined by enforcing a 2% logarithmic decrement on the first fore-aft, side-side, and torsional tower
modes (modes 1, 2, and 7, respectively). The final model-specific values can be found in the respective input files.
24
5 Nacelle, Drivetrain, and Hub
The IEA Wind 15-MW reference wind turbine uses a direct-drive layout with a permanent-magnet, synchronous,
radial flux generator in a simple and compact nacelle layout. Direct-drive wind turbine generators offer a number
of advantages over geared drivetrains, including fewer parts, lower complexity, higher reliability, and additional
flexibility in designing for special topologies. However, the direct coupling of the generator at very low speeds
requires large physical dimensions and higher mass, which incur transportation, assembly, and servicing challenges.
Countering these challenges involves an optimal balance of generator location, the number of bearings, internal
or external stator/rotor arrangement, rotor/stator inactive substructure geometries, ancillary component interfaces,
among other considerations [42].
(b)
(a)
Figure 5-1. A sketch and CAD model of the nacelle layout of the 15-MW direct-drive wind turbine. Not to scale
and some structural details omitted. Blades (not shown), hub, shaft, and generator rotor rotate.
5.1 Nacelle Overview

Figure 5-1a shows a simple direct-drive nacelle layout with an outer-rotor permanent-magnet generator. The assem-
bly consists of a hub shaft supporting the turbine and generator rotors on two main bearings housed on a stationary
turret that is cantilevered from the bedplate. The bedplate transmits the hub loads and weight of the rotor, generator,
hub, shaft, and turret to the tower. The three-dimensional CAD illustration is shown in Figure 5-1b. In determining
the optimal dimensions of the various components, the highest acceptable equivalent von Mises stress under extreme
aerodynamic loads measured during extreme turbulence conditions was 200 mega pascal (MPa). These ultimate
loads were assumed to represent multiaxial loading under these conditions, as suggested by International Elec-
trotechnical Commission (IEC) 61400-1. The full mass breakdown of the nacelle and its components is summarized
in Table 5-1.
5.2 Hub
The hub design is based on a simple spherical hub shell model with cutouts for the blades and the flange. The hub
diameter is calculated as a function of the blade root diameter. The thickness of the hub shell is designed to with-
25
Table 5-1. Lumped Masses and Moments of Inertia for the Nacelle Assembly
(The coordinate system has its origin at the tower top, with x pointed downwind [parallel to ground/water and not the shaft] and z pointed up)
Name XT T [m] ZT T [m] Mass [t] Ixx [kg m2 ] Iyy [kg m2 ] Izz [kg m2 ]
Yaw system 0.000 -0.190 100.0 490,266 490,266 978,125
Turret nose 5.786 4.956 11.394 12,571 10,890 10,909
Inner generator stator 5.545 4.913 226.629 3,777,313 2,012,788 2,042,312
Outer generator rotor 6.544 5.033 144.963 3,173,003 1,673,269 1,691,864
Shaft 6.208 5.000 15.734 33,009 22,906 23,018
Hub 10.604 5.462 190.0 1,382,171 2,169,261 2,160,637
Bedplate 0.812 2.697 70.329 398,973 515,880 535,055
Flange 4.593 4.831 3.946 4,081 2,065 2087
Misc. equipment 0.000 0.500 50.0 16,667 16,667 25,000
TDO shaft bearing 6.582 5.040 2.230 3,515 1,784 1,803
SRB shaft bearing 5.388 4.914 5.664 8,930 4,593 4,641
Nacelle total 5.486 3.978 820.888 12,607,277 21,433,958 18,682,468

Nacelle total minus hub 3.945 3.352 630.888 10,680,747 122,447,810 10,046,187
kg m2 kilogram square meters m meters t metric tons
TDO tapered double outer SRB spherical roller bearing
stand the forces generated in an emergency shutdown event. The hub flange diameter is half of the hub diameter, and
the flange thickness is four times the thickness of the hub shell. Edges of the flange are designed to minimize stress
concentrations. The pitch system is not modeled in extensive detail and its mass is estimated using regression fits.
5.3 Main Shaft, Bearings, and Turret

A depiction of the main shaft, bearings, and turret is shown in Figure 5-2, with key parameters and dimensions listed
in Table 5-2. The main shaft has a hollow cylindrical cross section, with a constant wall thickness and a tilt angle of
6◦ . The main shaft, along with the rotor, is supported by two main bearings. Both these main bearings have rotating
outer raceways and fixed inner raceways. The outer raceways and bearing housing are accommodated by a turret
held by the bedplate. The entire weight of the turbine rotor, generator rotor, and hub loads are transmitted by the
main shaft to the turret via the bearings.
Figure 5-2. CAD illustration of (a) the main shaft and (b) turret (also called the nose); dimensions are documented in Table 5-2
The inner diameter of the main shaft was designed with sufficient clearance for the nacelle turret and the main
bearings. The thickness of the shaft was determined by ensuring a safe load path from the rotor while limiting the
maximum deflection at the generator. It is assumed that the entire thrust load is transmitted by the main shaft to the
upwind main bearing.
26
The turret also has a hollow cylindrical cross section, with a constant wall thickness cantilevered from the bedplate.
The inner diameter was set to 2 m to provide clearance for technician access. The thickness of the turret cylinder
was determined by computing the reactions from the main bearings while limiting the maximum radial deflections at
the interface with the generator rotor and stator. It is assumed that the turret carries all of the thrust imparted by the
upwind main bearing, the moments from the transverse forces, and the torque.
The paired set of bearings consists of a fixed upwind bearing and floating downwind bearing. A tapered double
outer configuration was chosen for the locating bearing and a spherical roller bearing for the nonlocating bearing.
The chosen bearing solutions were based on recommendations found in [43], with a basic lifetime estimation for
ultimate loads, as described in the DNV guidelines. Loads from the aeroelastic simulations were used to determine
the reactions at the main bearings. The upwind bearing was assumed to carry all axial loads and moments, whereas
both the upwind and downwind bearings share the radial forces. These forces were directly transmitted to the turret
and from there to the bedplate and tower. Because the downwind bearing carries smaller loads, we propose a bearing
with a lower load rating.
Table 5-2. Main Shaft Dimensions, Bearing, and Loads Used in Sizing
Symbol Description Value Units
Lfl Hub flange length 0.358 m

Lmb1 Distance of upwind bearing 1.0 m
Lgr Generator rotor location 0.1 m
L12 Distance between upwind and downwind bearings 1.2 m
Las Main shaft center of mass 1.25 m
γ Shaft tilt angle 6.0 deg
Lms Length of main shaft 2.2 m
Rosh Outer radius of the main shaft 3.0 m
Rish Inner radius of the main shaft 2.8 m
Lgs Location of generator stator from bedplate flange 0.25 m
Lan Turret center of mass 1.2 m
L2n Distance of downwind bearing from bedplate flange 0.9 m
Lan Turret length 2.2 m
Ron Outer radius of the turret 2.2 m
Rin Inner radius of the turret 2.0 m
Mb1 Tapered double outer ring: mass 2,230 kg
Mb2 Spherical roller bearing: mass 5,664 kg
deg degrees m meters kg kilograms
5.4 Bedplate
The bedplate is a hollow, elliptically curved, cantilever beam with circular cross sections. The bedplate has a smaller
cross section that interfaces with the bedplate flange and a larger cross section that interfaces with the yaw bearing
at the tower top. The thickness of the circular ring elements is optimized to constrain the cumulative stresses from
bending, torsion, shear forces, and axial loads to 200 MPa. The total end deflection is an input into the turret de-
sign for an accurate air-gap deflection estimate. The driving load case is taken from the extreme aerodynamic and
turbulent wind field. Figure 5-3 shows the CAD illustration of the bedplate design based on the critical dimensions
provided in Table 5-3.
5.5 Yaw System

The yaw system bearings are based on those available in SKF’s catalog, specifically double-row, angular, contact ball
bearings. The yaw system mates the bedplate base with the tower top at a diameter of 6.5 m.
27
(a)
(b)
Figure 5-3. A CAD illustration of the bedplate

.
Table 5-3. Bedplate Properties
Ltop Tower top distance from turret flange 5 m

Hnose Nose height 4.875 m
Dtop Tower top diameter 6.5 m
t Bedplate wall thickness 50 mm
Mbed plate Bedplate mass 70,329 kg
kg kilograms m meters mm millimeters
28
5.6 Direct-Drive Generator
The DrivetrainSE module within WISDEM [44] was used to design and optimize a 15-MW direct-drive synchronous
generator. The generator construction features an external rotor radial flux topology machine with a surface-mounted
permanent magnet (shown in Figure 5-4), with the design and performance parameters listed in Table 5-4. The
outer rotor layout facilitates a simple and rugged structure, easy manufacturing, short end windings, and better heat
transfer between windings and teeth than the inner rotor configuration. The stator design features fractional slot
layout double-layer concentrated coils, which maximize the fundamental winding factor. The design optimization
followed the guidance found in [45–47]. The guiding assumptions in the design include the following:
• The rotor magnets comprise N40-grade sintered neodymium (NdFeB) magnets, with a remnant flux density of
1.28 tesla (T) and relative permeability of 1.06.
• Only radial components of the air-gap flux are considered, meaning leakage and fringing effects are neglected.
• A slot-pole combination of 2–5 was chosen to derive the concentrated winding layout using Cro’s technique
and winding vector method [45].
• The magnet width was assumed to be 80% of the pole pitch.
• The design is symmetric and the magnetic circuit is not saturated.
• The magnetic and electrical loadability are determined for a typical tangential stress of 60 kPa. The specific
current loading is assumed to be less than 100 kilo-ampere (kA) per meter, provided that an efficient thermal
management system is in place so that the output power of the machine can be improved without increasing its
size.
• Single-sided disc-type support structures are assumed to hold the electromagnetically active materials to
greatly simplify the assembly.
• The air-gap length is assumed to be 1/1,000th of the air-gap diameter. The air gap is allowed to deflect by up to
20%, with the support structure, shaft, and turret contributing to the deformation.
The optimized design features 100 pole pairs dispersed in an air-gap diameter of 10.53 m, a stack length of 2.17 m,
and an efficiency of 96.5%. The generator weighs 372 t, split between 227 t for the stator and 145 t for the rotor. At
least 50% of the mass stems from the structural support.
5.7 Nacelle Damping

We chose the stiffness-proportional damping for the shaft to produce 5% of critical modal damping for the free-free
mode of the generator/rigid-rotor system. This value was chosen based on the value in the NREL 5-MW reference
model. In particular, it can be shown that the stiffness proportional term, β , should be chosen such that:
s
Igen Irot
β = 2ζ , (5.1)
KDT (Igen + Irot )
where ζ is the desired modal damping, Ii is the generator or rotor inertia, and KDT is the equivalent stiffness of
the drivetrain. For this turbine, the rigid rotor inertia is approximately 3.524 605 × 108 kilogram-square meters
(kgm2 ), calculated from the HAWC2 blade structural file while ignoring coning and prebend. This results in
β =4.457 544 × 10−4 for the torsional motion.
29
(a) CAD layout
(b) Design parameters
Figure 5-4. A CAD illustration of an outer rotor direct-drive generator with electromagnetic and structural design parameters
30
Table 5-4. Electromagnetic and Structural Design of the 15-MW Direct-Drive Generator
Pr Rated power at generator terminals 15 MW

ωr Rated speed 0.792 rad/s
fe Electrical frequency 12.6 Hz
Tr Rated torque 21.03 MN m
Electromagnetic Design
rg Air-gap radius 5.08 m

l Core length 2.17 m
g Air-gap length 10.16 mm
p Poles 200 -
S Stator slots 240 -
hys Stator yoke thickness 46.52 mm
hyr Rotor yoke thickness 63.62 mm
τp Pole pitch 159.55 mm
τs Slot pitch 132.7 mm
hs Slot height 400.03 mm
bs Slot width 57.3 mm
bt Tooth width 74.85 mm
hm Magnet height 58.39 mm
bm Magnet width 127.64 mm
V RMS line voltage 4770.34 V
I Nominal winding current (RMS) 1084.55 A
Rs Stator winding resistance per phase 0.16 Ω
Ns Stator winding turns per phase 320 -
B̂g Peak air-gap flux density 1.06 T
Js Winding current density 3.39 A/mm2
A1 Specific current loading 92.46 kA/m
η Efficiency at full load 96.55 %
MIron Iron mass 180.95 t
MCu Copper mass 9.01 t
MMagnet Magnet mass 24.20 t
MActive Total active mass 214.16 t
Structural Design
hss Stator rim thickness 46.59 mm

ts Stator disc thickness 79.97 mm
hsr Rotor rim thickness 63.69 mm
tr Rotor disc thickness 81.75 mm
δl Rotor stator clearance 825 mm
MSStru Total rotor structural steel mass 86.2 t
MRStru Total stator structural steel mass 71.1 t
MGen Estimated total generator mass 371.57 t
MW megawatts V volts
rad/s radians per second A amperes
Hz hertz (cycles per second) Ω ohms
MN m mega-newton meters T tesla
m meters A/mm2 amperes per square millimeter
mm millimeters kA/m kilo-amperes per meter
t metric tons
31
6 Load Assessment
An IEC design load case (DLC) [4] analysis study was conducted to determine the worst-case ultimate loading on
key design constraining components. Table 6-1 provides details on the DLC cases simulated in OpenFAST and Table
6-2 specifies the corresponding metocean inputs. The maximum bending moments and deflections across all cases
were ranked, as shown in Figure 6-1 and Figure 6-2. Yaw-misaligned parked conditions with extreme wind speeds
and extreme coherent gust with a direction change result in the worst-case loading for this design. The worst-case
out-of-plane tip deflection is 22.8 m, leaving more than sufficient tower clearance, with an unbent blade tip-to-tower
clearance of 30.0 m. This margin suggests that the blade design is conservative and further aeroelastic optimization
could potentially improve the aerodynamic performance or cost of energy while still remaining within recommended
safety margins. A full fatigue analysis of this blade was not conducted, which could potentially be an issue for the
edgewise blade bending moments for blades of this size.
Table 6-1. Summary of IEC DLC Settings
Wind Wind Additional # of # of

DLC Condition Speeds Settings Seeds Simulations
1.1 NTM 3:2:25 m/s - 6 72

1.3 ETM 3:2:25 m/s - 6 72
1.4 ECD Vr , Vr ±2 m/s ± Dir. Change - 6
1.5 EWS 3:2:25 m/s ± Vert./Horz. - 48
6.1 EWM V50 Yaw ±8° 6 12
6.3 EWM V1 Yaw ±20° 6 12
NTM normal turbulence model
ETM extreme turbulence model
ECD extreme coherent gust with direction change
EWS extreme wind shear
EWM extreme wind speed model
Vr = 10.8 m/s rated wind speed
V50 = 50.0 m/s 10-min average extreme wind speed with a 50-year return period
V1 = 40.0 m/s 10-min average extreme wind speed with a 1-year return period
Table 6-2. Metocean Conditions Used in DLC Analysis
Operation Wind Speed Significant Wave Height Peak Spectral Period

[m/s] [m] [s]
Normal 4 1.102 8.515

6 1.179 8.310
8 1.316 8.006
10 1.537 7.651
12 1.836 7.441
14 2.188 7.461
16 2.598 7.643
18 3.061 8.047
20 3.617 8.521
22 4.027 8.987
24 4.516 9.452
Extreme 1-yr return 40 9.686 16.654
Extreme 50-yr return 50 11.307 18.505
m meters m/s meters per second s seconds
32
Blade Root Blade Root Tower Base Tower Base
Moment, Moment, Moment, Moment,
Flapwise Edgewise Side-Side Fore-Aft
400
Moments [MN-m]
300
200
100
0
.5 .4 .5 .5 .4 .4 .5 .1 .3 .1 .3 .1 .3 .3 .3 .4 .4 .3 .1 .3
L C 1 LC 1 LC 1 LC 1 LC 1 L C 1 LC 1 LC 1 LC 1 LC 1 LC 6 LC 6 LC 6 LC 6 LC 6 LC 1 LC 1 LC 1 LC 1 LC 1
D D D D D D D D D D D D D D D D D D D D
Figure 6-1. DLC ranking of maximum blade root and tower base bending moments
Blade Tip Blade Tip Tower Top Tower Top

Flapwise Edgewise Side-Side Fore-Aft
20
Deflections [m]
15
10
5
0
.4 .5 .4 .5 .5 .1 .3 .3 .1 .3 .1 .3 .3 .1 .3 .4 .4 .3 .4 .1
L C 1 LC 1 LC 1 LC 1 LC 1 L C 6 LC 6 LC 6 LC 6 LC 6 L C 6 LC 6 LC 6 LC 6 LC 6 L C 1 LC 1 LC 1 LC 1 LC 1
D D D D D D D D D D D D D D D D D D D D
Figure 6-2. DLC ranking of maximum blade tip and tower top deflections
33
6.1 Conclusions
Publicly available reference wind turbine designs have proven to be an invaluable tool to the wind energy research
community. As the technology has continued to mature and improve, subsequent reference turbines [2, 3, 5–9]
have been developed to keep pace with, and ahead of, industry trends. In order to be relevant for forward-looking
technology development and analysis efforts, reference wind turbines need to represent state-of-the-art or near-future
wind turbine technology. This work was motivated by a gap in the available offshore reference turbine designs,
between 10 MW and 20 MW, while the industry is moving beyond 10-MW designs.
This report documents the design and performance of the IEA Wind 15-MW reference wind turbine, jointly devel-
oped by NREL, DTU, and UMaine, in coordination with the second work package in IEA Wind Task 37 on Wind
Energy Systems Engineering. This reference wind turbine represents a standardized design that can be used for
concept studies as the industry progresses toward larger machines. It is a conventional three-bladed upwind design
with a rotor diameter of 240 m; a 150-m hub height; a variable-speed, collective pitch controller; and a low-speed,
direct-drive generator. A baseline steel monopile support structure is presented, which has been sized at a 30-m water
depth. Subsequent work, in collaboration with UMaine, will document a floating support structure configuration on a
steel semisubmersible.
Further details of the design, including OpenFAST, HAWC2, and WISDEM input files, are available at the IEA
Wind Task 37 GitHub.
34
References
[1] WindEurope. Offshore Wind in Europe. Key Trends and Statistics. Tech. rep. WindEurope, 2019.
[2] Pietro Bortolotti, Helena Canet Tarrés, Katherine Dykes, Karl Merz, Latha Sethuraman, David Verelst, and
Frederik Zahle. IEA Wind TCP Task 37: Systems Engineering in Wind Energy - WP2.1 Reference Wind Tur-
bines. Tech. rep. International Energy Agency, 2019. DOI: 10.2172/1529216.
[3] INNWIND.EU. http://www.innwind.eu/publications/deliverable-reports. 2017.
[4] International Electrotechnical Commission. Wind Turbines – Part 1: Design requirements. Tech. rep. IEC
61400-1 Ed.3. 2005.
[5] G. Bywaters, V. John, J. Lynch, P. Mattila, G. Norton, J. Stowell, M. Salata, O. Labath, A. Chertok, and D.
Hablania. Northern Power Systems WindPACT Drive Train Alternative Design Study Report. Tech. rep. 2004.
[6] J. Jonkman, S. Butterfield, W. Musial, and G. Scott. Definition of a 5-MW Reference Wind Turbine for Off-
shore System Development. Tech. rep. NREL/TP-500-38060. National Renewable Energy Laboratory, 2009.
[7] C. Bak, F. Zahle, R. Bitsche, and T. Kim. Description of the DTU 10MW reference wind turbine. Tech. rep.
DTU Wind Energy Report I-0092. DTU Wind Energy, 2013.
[8] C. Desmond, J. Murphy, L. Blonk, and W. Haans. “Description of an 8 MW reference wind turbine.” In:
Journal of Physics: Conference Series (2016). DOI: 10.1088/1742-6596/753/9/092013.
[9] D. Griffith and T. Ashwill. The Sandia 100-meter All-glass Baseline Wind Turbine Blade : SNL100-00. Tech.
rep. SAND2011-3779. Sandia National Laboratories, 2011.
[10] Katherine Dykes, Peter Graf, George Scott, Andrew Ning, Ryan King, Yi Guo, Taylor Parsons, Rick Damiani,
Fort Felker, and Paul Veers. “Introducing WISDEM: An Integrated System Model of Wind Turbines and
Plants.” In: Third Wind Energy Systems Engineering Workshop, January 14, 2015. Boulder, CO, 2015.
[11] WISDEM. 2019. URL: https://github.com/WISDEM/WISDEM.
[12] “Aero-Elastic Optimization of a 10 MW Wind Turbine.” English. In: Proceedings - 33rd Wind Energy Sympo-
sium. Vol. 1. American Institute of Aeronautics and Astronautics, 2015, pp. 201–223.
[13] “Design of an aeroelastically tailored 10 MW wind turbine rotor.” English. In: vol. 753. IOP Publishing, 2016.
DOI : 10.1088/1742-6596/753/6/062008.
[14] S. A. Ning. CCBlade Documentation: Release 1.1.0. Tech. rep. National Renewable Energy Laboratory
(NREL), Golden, CO (United States), 2013. URL: http://wisdem.github.io/CCBlade/.
[15] SA Ning. RotorSE Documentation: Release 0.1.0. Tech. rep. National Renewable Energy Lab. (NREL),
Golden, CO (United States), 2014. URL: http://wisdem.github.io/RotorSE/.
[16] Gunjit S. Bir. User’s Guide to PreComp. Tech. rep. Golden, CO: National Renewable Energy Laboratory,
2005.
[17] José Pedro Albergaria Amaral Blasques. User’s Manual for BECAS: A cross section analysis tool for anisotropic
and inhomogeneous beam sections of arbitrary geometry. Tech. rep. 2012.
[18] G. Bir. User’s Guide to BModes. Tech. rep. NREL/TP-500-39133. Golden, Colorado: National Renewable
Energy Laboratory, 2005.
[19] Jason M. Jonkman and Marshall L. Buhl Jr. FAST User’s Guide. Tech. rep. NREL/EL-500-38230. Golden,
Colorado: National Renewable Energy Laboratory, 2005. DOI: 10.2172/15020796.
[20] OpenFAST Documentation. 2019. URL: https://openfast.readthedocs.io/en/master/.
[21] Torben J. Larsen and Anders Melchior Hansen. How 2 HAWC2, the user’s manual. Tech. rep. 2007.
[22] M. Hansen, L. Henriksen, C. Tibaldi, L. Bergami, D. Verelst, G. Pirrung, and R. Riva. HAWCStab2 User
Manual. Tech. rep. Roskildem, Denmark: Technical University of Denmark, 2018.
[23] Justin S. Gray, John T. Hwang, Joaquim R. R. A. Martins, Kenneth T. Moore, and Bret A. Naylor. “OpenM-
DAO: An open-source framework for multidisciplinary design, analysis, and optimization.” In: Structural and
Multidisciplinary Optimization 59.4 (Apr. 2019), pp. 1075–1104. DOI: 10.1007/s00158-019-02211-z.
[24] Gordon M. Stewart, Amy Robertson, Jason Jonkman, and Matthew A. Lackner. “The creation of a compre-
hensive metocean data set for offshore wind turbine simulations.” In: Wind Energy 19.6 (2016), pp. 1151–
1159. DOI: 10.1002/we.1881.
[25] Jess A Michelsen. Basis3D-a platform for development of multiblock PDE solvers. Tech. rep. AFM 92-05,
Technical University of Denmark, 1992.
[26] Jess A Michelsen. “Block structured Multigrid solution of 2D and 3D elliptic PDE’s.” PhD thesis. 1994.
[27] Niels N. Sørensen. “General purpose flow solver applied to flow over hills.” Published 2003. PhD thesis.
1995.
35
[28] Niels N. Sørensen. HypGrid2D a 2-D mesh generator. Tech. rep. 1035(EN). 1998.
[29] Florian Menter. “Zonal two equation kw turbulence models for aerodynamic flows.” In: 23rd fluid dynamics,
plasmadynamics, and lasers conference, p. 2906.
[30] Mark Drela and Michael B Giles. “Viscous-inviscid analysis of transonic and low Reynolds number airfoils.”
In: AIAA journal 25.10 (1987), pp. 1347–1355.
[31] Zhaohui Du and Michael S. Selig. “A 3-D stall-delay model for horizontal axis wind turbine performance
prediction.” English (US). In: AIAA/ASME Wind Energy Symposium, 1998 ; Conference date: 12-01-1998
Through 15-01-1998. Jan. 1998, pp. 9–19.
[32] K. Dykes et al. System Modeling Frameworks for Wind Turbines and Plants. Tech. rep. In preparation. Inter-
national Energy Agency.
[33] Yu Wenbin. VABS Manual for Users. Tech. rep. West Jordan, UT: AnalySwift, 2011.
[34] T. Pflumm, W. Garre, and M. Hajek. “A preprocessor for parametric composite rotor blade cross-sections.” In:
44th European Rotocraft Forum. Delft, The Netherlands, 2018.
[35] José Pedro Albergaria Amaral Blasques and Mathias Stolpe. “Multi-material topology optimization of lam-
inated composite beam cross sections.” In: Composite Structures 94.11 (2012), pp. 3278–3289. ISSN: 0263-
8223. DOI: https://doi.org/10.1016/j.compstruct.2012.05.002.
[36] J. Rinker, E. Gaertner, P. Bortolotti, W. Skrzypinski, F. Zahle, H. Bredsmoe, G. Barter, and K. Dykes. “Com-
parison of loads from HAWC2 and FAST for the IEA 15 MW Reference Wind Turbine.” In: Torque 2020.
[37] National Renewable Energy Laboratory. ROSCO. Version 0.1.0. 2019. URL: https://github.com/NREL/rosco.
[38] M. H. Hansen and L. C. Henriksen. Basic DTU Wind Energy controller. Tech. rep. DTU Wind Energy Report
E-0018. DTU Wind Energy, 2013.
[39] N. J. Abbas, L. Pao, and A. Wright. “An Update to the National Renewable Energy Laboratory Wind Turbine
Controller.” In: Journal of Physics: Conference Series (). In preparation.
[40] Laszlo Arany, S. Bhattacharya, John Macdonald, and S. J. Hogan. “Design of monopiles for offshore wind
turbines in 10 steps.” In: Soil Dynamics and Earthquake Engineering 92 (2017), pp. 126–152. ISSN: 0267-
7261. DOI: https://doi.org/10.1016/j.soildyn.2016.09.024. URL: http://www.sciencedirect.com/science/article/
pii/S0267726116302937.
[41] Suresh C. Arya, Michael W. O’Neill, and George Pincus. Design of structures and foundations for vibrating
machines. Gulf Pub Co, 1979.
[42] Johan N. Stander, Gerhard Venter, and Maarten J. Kamper. “Review of direct-drive radial flux wind turbine
generator mechanical design.” In: Wind Energy 15.3 (2012), pp. 459–472. DOI: 10.1002/we.484.
[43] E. Smith. “DESIGN AV NACELLE FOR EN 10 MW VINDTURBIN.” MA thesis. Norwegian university of
science and technology, 2012.
[44] Latha Sethuraman and Katherine Dykes. GeneratorSE: A Sizing Tool for Variable-Speed Wind Turbine
Generators. Tech. rep. NREL/TP-5000-66462. National Renewable Energy Laboratory, 2017. URL: https :
//www.nrel.gov/docs/fy17osti/66462.pdf.
[45] Hung Vu Xuan. “Modeling of Exterior Rotor Permanent Magnet Machines with Concentrated Windings.”
PhD thesis. 2012.
[46] Ion Boldea. Synchronous generators. CRC Press, 2015.
[47] Alasdair Stewart McDonald. “Structural analysis of low speed, high torque electrical generators for direct
drive renewable energy converters.” PhD thesis. University of Edinburgh, 2008.
[48] P. C. J. Hoogenboom and R. Spaan. “Shear Stiffness and Maximum Shear Stress of Tubular Members.” In:
Proceedings of the Fifteenth International Offshore and Polar Engineering Conference. International Society
of Offshore and Polar Engineers. (Jan. 1, 2005). ISOPE-I-05-399. Seoul, Korea, 2005.
36
A Blade Cross Sections
Renderings of the spanwise blade cross sections are shown in this section. The 70% spanwise location is shown in
Figure 2-6.
Figure A-1. Blade cross section at 0% span
37
38
39
40
B Tower and Soil Modeling
B.1 Tower Data

The equations used to calculate cross-sectional tower properties are provided in this section. The definitions of the
terms used in Figure 4-3 are:
• Mass density is equal to m
2
• Flapwise inertia is equal to m rix
2
• Edgewise inertia is equal to m riy
• Flapwise stiffness is equal to EIx
• Edgewise stiffness is equal to EIy
• Torsional stiffness is equal to GK
• Axial stiffness is equal to EA.
B.1.1 Station, r [m]
The distance along the tower from the base. For the land-based HAWC2 model, this is assumed to start at 0 meters
(m) and end at 145.0 m. Thus, the tower model includes the transition piece, which extends from 0 m to 15 m.
B.1.2 Mass per Unit Length, m [kg/m]
Mass per unit length of the tower:
m=ρA (B.1)
= ρ π (D/2)2 − (D/2 − t )2 ,

(B.2)
where A is the cross-sectional area, D is the outer diameter, and t is the wall thickness at a given station.
B.1.3 Center of Mass, xm [m]
The x location of the center of mass. Because the tower is axisymmetric, this value is zero, xm = 0.
B.1.4 Center of Mass, ym [m]
The y location of the center of mass. Because the tower is axisymmetric, this value is zero, ym = 0.
B.1.5 Radius of Gyration, rix [m]
The radius of gyration around the principal bending axis, xe . We use the radius of gyration to calculate the mass
moment of inertia for a cross section, which we need for inertia calculations. for an isotropic circular tube:
r
Ix
rix = , (B.3)
A
s
π [(D/2)4 − (D/2 − t )4 ] /4
= , (B.4)
π [(D/2)2 − (D/2 − t )2 ]
r
1
= [(D/2)2 + (D/2 − t )2 ]. (B.5)
4
41
B.1.6 Radius of Gyration, riy [m]
Radius of gyration around the principal bending axis, ye , is the same as rix due to symmetry.
B.1.7 Shear Center, xs [m]
The x-coordinate of the shear center. Because the cross section is symmetric about both x and y, the shear center is
collocated with the elastic center (which is at the origin), xs = 0.
B.1.8 Shear Center, ys [m]
The y coordinate of the shear center. Because the cross section is symmetric about both x and y, the shear center is
collocated with the elastic center (which is at the origin), ys = 0.
B.1.9 Young’s Modulus, E [Pa]
The Young’s modulus with values in Table 4-1.
B.1.10 Shear Modulus, G [Pa]
The shear modulus with values in Table 4-1.
B.1.11 Area Moment of Inertia, Ix [m4 ]
Area moment of inertia around the principal bending axis, xe :

Z
Ix = x2 dxdy (B.6)
A
π
(D/2)4 − (D/2 − t )4

= (B.7)
4
B.1.12 Area Moment of Inertia, Iy [m4 ]
Area moment of inertia around the principal bending axis, ye :

Z
Iy = y2 dxdy (B.8)
A
π
(D/2)4 − (D/2 − t )4

= (B.9)
4
B.1.13 Torsional Stiffness Constant, K [m4 /rad]
Torsional stiffness constant calculated about the z axis at the shear center. Because we assume a circular section, this
is equivalent to the polar moment of inertia:
Z
K= r2 dxdy, (B.10)
A
π
(D/2)4 − (D/2 − t )4 .

= (B.11)
2
B.1.14 Shear Reduction Factor, kx [-]
Shear factor, also called the shear reduction factor, for shear in the x direction. Per [48], we use the following shear
factor:
1 3 2t
kx = + . (B.12)
2 4D
42
B.1.15 Shear Reduction Factor, ky [-]
Shear factor, also called the shear reduction factor, for shear in the y direction. Per [48], we use the following shear
factor:
1 3 2t
ky = + . (B.13)
2 4D
B.1.16 Cross-Sectional Area, A [m2 ]
The area of the cross section:

Z
A= dxdy, (B.14)
A
= π (D/2)2 − (D/2 − t )2 .

(B.15)
B.1.17 Structural Pitch, θs [deg]
This is the angle between x and the principal bending axis most parallel to x. Because the tower is axisymmetric, this
is zero, θs = 0).
B.1.18 Elastic Center, xe [m]
The x location of the elastic center, which is the intersection point for the principal bending axes. Because the tower
is axisymmetric about z, this is 0, xe = 0.
B.1.19 Elastic Center, ye [m]
The y location of the elastic center. Because the tower is axisymmetric about z, this is 0, ye = 0.
43
B.2 Soil Model
B.2.1 Vertical Stiffness
Stiffness of the soil in the vertical direction for circular structures. This is sometimes ignored because monopiles and
towers are relatively stiff in their axial directions:
h
ηz = 1 + 0.6 (1 − ν ) (B.16)
r0
4Gr0
kz = ηz (B.17)
1−ν
B.2.2 Horizontal Stiffness
Stiffness of the soil in the lateral direction for circular structures:

h
ηx = 1 + 0.55 (2 − ν ) (B.18)
r0
32 (1 − ν ) Gr0
kx = ηz (B.19)
7 − 8ν
B.2.3 Rocking Stiffness
Stiffness of the soil in the rocking direction for circular structures:

3
h h
ηψ = 1 + 1.2 (1 − ν ) + 0.2 (2 − ν ) (B.20)
r0 r0
8Gr03
kψ = ηψ (B.21)
3 (1 − ν )
B.2.4 Torsional Stiffness
Stiffness of the soil in the rocking direction for circular structures:
16Gr03
kφ = (B.22)
3
44

March 2020: IEA Wind TCP Task 37

Uploaded by

Copyright:

Available Formats

March 2020: IEA Wind TCP Task 37

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

March 2020: IEA Wind TCP Task 37

Uploaded by

Copyright:

Available Formats

March 2020

IEA Wind TCP Task 37

Definition of the IEA Wind

NREL is a national laboratory of the U.S. Department of Energy Technical Report

This report is available at no cost from the National Renewable

U.S. Department of Energy (DOE) reports produced after 1991

NREL prints on paper that contains recycled content.

Name Institution Contribution

Overview and Motivation

Blade and Rotor Properties

Tower and Monopile Properties

Nacelle and Drivetrain Properties

Parameter Units Value

Control - Variable speed

Rotor diameter m 240

Drivetrain - Direct drive

Transition piece height m 15

Description Value Units

Blade length 117 m

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

4 Tower and Monopile Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

5 Nacelle, Drivetrain, and Hub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

A Blade Cross Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

B Tower and Soil Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Figure ES-1.The IEA Wind 15-MW reference wind turbine . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Table ES-1.Key Parameters for the IEA Wind 15-MW Turbine . . . . . . . . . . . . . . . . . . . . . . . . . . iv

1.1 The Role of Reference Wind Turbines

1.2 Overall Turbine Parameters

1.3 Design Tools and Methodologies

Parameter Units DTU 10-MW Turbine IEA Wind 15-MW Turbine

Rotor diameter m 178.3 240

Drivetrain - Medium speed Low speed

Shaft tilt angle deg 5 6

Transition piece height m 10 15

Role NREL Tool Chain DTU Tool Chain

System Design WISDEM [10], [11] HAWTOpt2 [12], [13]

1.4 Model Availability

1.5 Meteorological Ocean Environment

Table 2-1. Blade Properties

Description Value Units

Blade length 117 m

2.1 Blade Aerodynamic Properties

2.2 Blade Structural Properties

1.0 FFA-W3-241 Re=1e7

Relative Thickness [%]

Figure 2-4. Blade planform spanwise quantities

Figure 2-6. Blade cross section at 70% span

3.1 Controller Properties

3.1.1 Controller Methodology

Region Reg. Region

Rotor Thrust [MN]

Region Reg. Region

Tip Speed Ratio, [-]

Figure 3-2. CCBlade steady-state blade element momentum aerodynamic power

Figure 3-3. Spanwise forces on the blade as a function of wind speed

Table 4-1. Material Properties for the Tower

Parameter Symbol Value Units

Young’s modulus, steel E 2 00E11 Pa

6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 20 25 30 35 40 45 50 55