www.ecn.nl
Wind Energy
University of Utrecht, November 19, 2018
Guest lecture for GEO4-2503 ENSM-Energy Conversion Technologies II Gerard Schepers
Foto NLR
Education
• 1986:Delft University of Technology, Department of Aerospace Technology, graduated at High Speed Aerodynamics
• 2000: Propadeuse in Mathematics teaching
• PhD thesis completed and defended on November 27th2012 “Engineering models in wind turbine aerodynamics’
Appointments:
• Researcher at the Energy Research Center of the Netherlands (ECN), 1986-present Since April 1st2018: ECN part of TNO
• Summer 2010 and 2011: visiting professor at Korean Universities
• Since 2012: HBO professor at University of Applied Sciences Leeuwarden UAS Hanze
Main Topics:
• Rotor Aerodynamics, Windfarm Aerodynamics, Aeroelasticity, Aeroacoustics, Wind Tunnel Testing, Wind Turbine Blade Design.
• Researcher in various national/international research projects and coordinator of 7 joint European research projects and 5 joint mondial IEA research projects
• Involved in various industry related projects.
• December 2007-December 2009: Temporary stationed at Suzlon Blade Technology for consultancy activities during 4 days/month
• Contributing as a (guest) lecturer to annual lecture series at e.g. Technical Universities of Eindhoven and Delft, University of Utrecht, WUR
Scientific journal papers and books:
• Author or co-author of more than 40 peer-reviewed scientific articles and more than 100 conference papers and technical reports and 2 book chapters
• Invited chairman and session editor at almost all major international wind energy conferences
Contents
Part 1
•
Introduction, Costs and contribution to energy demand,Global situation Wind energy, Challenges
•
Wind ResourcesPart 2:
•
Energy Conversion•
Design loads•
Technology•
Wind farm effects•
Operation and MaintenanceWhy (electricity from) wind energy
•
Environmentally friendlyClimate change
IEA World Energy Outlook 2011:
The door to 20C is
closing
By Presidencia de la República Mexicana - https://www.flickr.com/photos/presidenciamx/23430273715/,
2015 United Nations Climate Change
Conference: The door
Why (electricity from) wind energy
•
Environmentally friendly•
Unlimited (even shale gas is notunlimited….)
•
Secures independancy frompolitically unstable/unfriendly countries
“We have a fully automatic Minister of Foreign Affairs:
when he opens up his mouth, somewhere else in the world the oil supply is switched off”
Wim Kan (Dutch conferencier): about minister van der Stoel
Why (electricity from) wind energy
•
Environmentally friendly•
Unlimited•
Secures independancy from politicallyunstable/unfriendly countries
•
Cheap compared to conventional (finite) sources on thelong run
•
Employment (strong position of Dutch off-shore industryand research institutes)
Involvement Dutch sector in
offshore wind farms
BECAUSE IT IS FUN!
http://www.strandbeest.com/index.php
Costs and contribution to the
energy demand
Historical and Forecasted Onshore Wind LCOE and Learning Rates
Price reduction of (on-shore) wind energy
Ryan Wiser, Karen Jenni Joachim Seel, Erin Baker, Maureen Hand, Eric Lantz, Aaron Smith Forecasting Wind Energy Costs and Cost Drivers: The Views of the World’s Leading Expert IEA Wind Task 26, June 2016
Costs of wind energy (2014!)
Turbine plus installation costs per unit power
• on land: 1250 Euro/kW
• on sea: 1800-2000 Euro/kW (different ways to account for costs for grid connection)
Costs of energy, highly dependent on wind climate:
• On-shore: 7 – 12.2 Eurocents/kWh • Off-shore:
– 13.5 Eurocents/kWh (hwater< 10m)
– 15 Eurocents/kWh (hwater > 10m favourable location)
• ‘Conventional energy’: 5-6 Eurocents/kWh • PhotoVoltaic: ~ 13.4 Eurocents/kWh
GROW contract
signature of CEO’s and minister
Off-shore windenergie expensive?
June 2016: GROW project: Reduce costs of off-shore wind energy to 7 ct/kWh in……
7 ct/kWh in 2026 seemed a reasonable target
7.27 cent/kWh (excl 1.4 cent/kWh cable)!!!!
Cost reduction (o.a. through technological innovation)
December 2016, even cheaper……
March 2018:
No subsidy at
all!
So wind energy costs 5-6 cent per kWh
5-2-2019 20
How much is 1 kWh?
5-2-2019 21
• Power measurements time trial Tour de France 2016: 37.5 km • Won by Tom Dumoulin in 50 minutes; power 438 W
(0.438 kW)
50 minutes = 50/60 hour: Energy =0.438*50/60 = 0.365 kWh • Froome: Much less power (loser…) but longer period
of 51 minutes
So almost same amount of energy: 0.418*51/60 =0.355 kWh
So how much is 1 kWh (= 5 cent)?
1 day of hard work…
23 Denmark 17 Nov 2016 22:53 Wind: 3466 MW Demand: 3455 MW >100% !!!
Electricity demand met by wind energy
(IEA Wind Strategic plan, October 2018)25
Global installed wind power 539 GW (end 2017)
5-2-2019 26
o Electricity demand the Netherlands • Working days
• ~ 16 GW day time • ~ 11 GW night time
5-2-2019 27
Expected cumulative installed
capacity per region 2017-2022
The Netherlands
2017: Installed wind power (Source CBS)
• On-shore: 4240 MW • Off-shore: 958 MW
Energy Agreement (2013) has led to a boost in installed wind power o Target for 2020:
• 6 GW on-shore wind energy
• 14% of all energy should be renewable o Target for 2023:
• 4.4 GW off-shore
Wind farm (700 MW) Existing wind farm / Under construction
Appointed area Proposed area
Dutch offshore projects
@North Sea
• Offshore wind farms in operation – OWEZ (108MW, Vestas V90) – Amalia (120MW, Vestas V80) – Luchterduinen (129MW, Vestas V112) – Gemini (600MW, Siemens SWT3.6) 31 Gemini 600 MW Siemens SWT 3.6 Luchterduinen 129MW Vestas V112 OWEZ (108MW Vestas V90) Amalia (120MW Vestas V80)
2012-05 www.we-at-sea.org 32
Two off-shore wind farms operational since 2007 at NW Netherlands
2008 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Source: TUDelft Offshore Engineering, Wybren de Vries
Shell-NUON
36 x Vestas 3MW V90 108MW 10 – 18 km from the coast
18 m water depth Monopiles
total investment appr. M€ 260
OWEZ E-Connection Evelop 60 Vestas 2 MW V80 120 MW 23 km from coast 19 - 24 m water depth mono piles
total investment appr. M€ 380
Princess Amalia
Photo: Jos Beurskens (We@Sea)
Another one became operational in
2016
•
Luchterduinen wind farm:
129 MW (43 Vestas V112 (3MW) turbines)
Gemini is operational too
Gemini: 600 MW 60 km above Schiermonnikoog
• Energy agreement offshore wind
– Build every year: 700MW – Target 2045MW in 2023
– Borssele I&II – Borssele III&IV – Hollandse kust Zuid – Hollandse kust Noord
Plan offshore wind farms
upto 2023
Tender Capacity (MW) Operational 2015 700 2019 2016 700 2020 2017 700 2021 2018 700 2022 2019 700 202336 5-2-2019
New Government:
There will be another 6GW
IN SUMMARY: THE DUTCH
OFF-SHORE FARMS
Operational (957MW) OWEZ 2006 108MW Amalia 2008 120MW Luchterduinen 2015 129MW Gemini 2017 600MW Tender (3.5GW) Borssele 2016 1400MW HKZ 2017/18 1400MW HKN 2019 700MW Roadmap 2030 (6.1GW) HKW 1400MW Fryslan Islands 700MW IJmuiden ver 4000MWSource: Letter of minister to House of Representatives
Are the Borssele locations the best places for our wind farms???
Satellite picture of wake effects behind
Belgian wind farms (Courtesy C. Hasager)
38
What are the scenarios beyond 2030?
5-2-2019 39
What are the scenarios beyond 2030?
‘The future of the North Sea’, PBL 2018.
5-2-2019 40
Space: 60 GW = 17-26% of
CHALLENGES
© H.Seifert www.hs-bremerhaven.de 500 kW 300 kW 50 kW 600 kW 1,500 kW 0 20 40 60 80 100 120 140 1980 1985 1990 1995 2000 2005 2010 2,500 kW 5,000 kW 4,500 kW R o to r d ia m e te r, m Year Serial-WT Prototypes
© H.Seifert www.hs-bremerhaven.de 500 kW 300 kW 50 kW 600 kW 1,500 kW 0 20 40 60 80 100 120 140 1980 1985 1990 1995 2000 2005 2010 2,500 kW 5,000 kW 4,500 kW R o to r d ia m e te r, m Year Serial-WT Prototypes
1 10 100 1000 10000 100000 0 20 40 60 80 100 120 Rotor Diameter, m B la d e M a ss , kg 2012-05 www.we-at-sea.org 44
Upscaling hampered by (theoretical)) square cube law
(weight increases with D3 and power with D2)
Much progress in blade technology
2012-05 47
Some challenges: The new LM88.4 blade
The LM 88.4 blades for Adwen 180 meter
turbine
5-2-2019 48
The largest rotating machines on earth
Rotating systems are always difficult
50
http://www.youtube.com/watch?v=8z-SSumb2ZA&noredirect=1
Challenges: The very different scales in wind energy
(free to Finn Gunner Nielsen)
Mesoscale(10000 10 km, days-hours)
Farm scale(10-1 km, 20 min-20 s)
Rotor scale (200 m-50 m, 10-2 s)
52 5-2-2019
Wind field = U (y,z,t)
Wind power fluctuations of 1 wind turbine:
Pr = rated power (Several MW’s….)
53
Courtesy:
J. Peinke, Forwind
54
Is this worrying???
Circles: 10 minute Boxes: 30 minute
H. Holtinen: Design and operation of power systems with large amount of wind power, IEA Task 25 final report, VTT 2013
Standard deviation of wind power fluctuations (as percentage of installed capacity) decreases rapidly with increasing distance
Power to gas
(source NSE1)Power to Gas, concept 1
(Source NSE1)
Power to Gas, concept 2
(Source NSE1)
5-2-2019 58
• Wind turbines are fatigue machines
• (Microscopic) cracks grow gradually leading to failure
• Fatigue: driven by load variation and the number of these variations (cycles)
• The lifetime of a wind turbine is 20 years
• In 20 years a wind turbine experiences very many (large) load variations
60 μm
5-2-2019 59
• Heavy large wind turbine blade rotating in the earth gravity field periodic
variation of gravity loads
• Wind turbines operate at lower part of the atmosphere with gusts, turbulence, wind shear
• Wakes of upstream turbines with increased turbulence and shear
• Control actions, grid disturbances Hence:
• Blades are flexible to reduce the load fluctuations
(System) DYNAMICS are extremely important, but the dynamics of an entire wind turbine system is difficult to
comprehend
R. Nijssen
60 μm
Loads and flexibilities of a wind turbine
Apart from the technological challenges there are social-economic challenges
•
Wind energy is more expensive than conventional energysources (Although…..)
•
Noise•
Visual impact on-shore problems off-shore *)•
Shadow•
Bird killers?*) Mind the higher price for off-shore compared to on-shore
62
Acoustic vibration monitoring
WT-Bird (pictures ECN)
Triggering algorithm Triggered video
registration
WT BIRD, shooting tennis balls to simulate birds collissions
WT-Bird Testing and validation
offshore
Simulating bird impacts with biodegradable sand bags offshore
Moment of sand bag impact. Size and weight of sand bag representative of song bird.
Contents
Part 1
•
Introduction, Costs, Global situation Wind energy,Challenges
•
Wind ResourcesPart 2:
•
Energy Conversion•
Technology•
Wind farm effects•
Operation and MaintenanceSome fundamentals on
wind energy to start with
First: Never confuse Energy with
Power!!!!
Energy (Work):
Force * distance= [N-m] = Joule Power (P=Energy per second) :
Force * velocity = [N-m/s] = Joule/s = Watt
(Electrical) power and energy; Commonly used units Power: kW = 103 W MW = 106 W GW = 109 W TW = 1012 W Energy: 1 kWh = 1000 W* 3600 s = 3.6 MWs = 3.6 MJ 1 MJ = 106 J = 0.28 kWh 1 GJ = 109 J = 280 kWh 1 TJ = 1012 J = 280 103kWh 1 PJ = 1015 J 1 EJ = 1018 J = 280 109 kWh 68
Units of wind speed
•
Stop thinking in wind forces (Beaufort)•
Wind speed should be expressed in m/s!! 69 op er at io na l r an geThe power is a result of the rotorshaft torque
which is a result of the in-plane blade loads
Rotor shaft torque T, rotational speed W Power = TW W Wind speed T(orque) = F1 r1 + F2 r2 F2 r2 F1 r1
F1and F2are resultants of in plane forces
In-plane force distribution: fip(r)
0 R
ip blades
Torque =
f rdrMind the units: •Torque = Nm
•W = rad/s (if W is given in rpm it should be recalculated to rad/s) •Power = Nm/s = Watt!
www.ecn.nl
Wind Resources/Wind Conditions
72
73
Wind Energy Conversion System (WECS)
Wind turbine Transformer Utility grid Wind V(x,y,z,t) Ekin = ½ ρV2 P/A= ½ ρV3 Ekin = ½ ρV2 P/A= ½ ρV3 Pm = TW P e = V I cosΦ
Wind speed, wind energy and wind
power
Kinetic energy in the wind: Ekin = ½.m.V2
1 m
1 m2
The mass of 1 m3 air at sea level is 1.225 kg
V m/s
1 m3 of air
ρ = 1.225 kg/m3;
kinetic energy per unit volume: Ekin per volume = ½.ρ.V2 [J/m3]
77
Wind energy = kinetic energy = ½ ρ V2 [J/m3],
= energy per unit volume Wind power = energy per second =
energy per volume * volume per second ½ ρ V2 AV
wind speed V
volume per second =
A V m3/s Area A [m2] ρ = ‘density’ = mass/unit volume
78
Wind speed, wind energy and wind power
V m
1 m2
A
ρ = 1.22 kg/m3
wind power density =
wind energy per second through unit area (A = 1m2)
Pw = ½ ρ V2 V
*
1 = ½ ρ V3 [W/m2]V [m/s]
V ρ Wind power density Pw
4 1.22 39 6 1.22 131 6 1.00 108 8 1.22 312 8 1.00 256 10 1.22 610 V3 makes site selection essential!!
79
Air density decreases with height above sea level:
at sea level: ρ = 1.225 kg/m3 at 1500 m a.s.l. and T = 5 °C: ρ = 1.05 kg/m3 General: p is pressure in Pa T is temp in K R = 287 J/kg-K Note: [K] = 273.16 + [°C]
Variation of density with height a.s.l.
RT
p
=
80 Height a.s.l. T [°C] ρ [kg/m3] [m] 0 15.0 1.225 200 13.7 1.202 500 11.75 1.167 1000 8.5 1.112 1500 5.25 1.058 2000 2.0 1.007
Standard ‘laps rate’: dT/dz = 0.65K/100m
81
The wind in Europe
Source: RISO/DTU
82 5-2-2019
The wind as source of energy
0 2 4 6 8 10 12 14 16 18 3 5 7 9
Annual average wind speed at 50m height [m/s] G en er at io n c o st i n € ce n ts /k W h 40m hub height 55m hub height 75 hub height Economical tower height ~ 1.1 Rotor diameter Fax Mtower
Long term mean wind speed at h=10m
Let’s anyhow make the tower height >
0.5 rotor diameter!!!
Tower height/Rotor diameter
5-2-2019 84
• Generally there is a trend
towards taller towers
• But not always!
– EWT-DW61turbine
(https://ewtdirectwind.com/):Tower height/
Rotor diameter = 0.8
Recommended Literature
• H.J.M. Beurskens et al Converting Offshore Wind into electricity The
Netherlands Contribution to offshore wind knowledge, Eburon Academic Publishers, 2011
• Twidell and Weir Chapters 7 (Wind Resource) and chapter 8 (Wind Power
Technology) from Renewable Energy Resources, Routedge, 2015
• J.F. Manwell, J.G. McGowan, A.L. Rogers Wind Energy Explained, Wiley
• For your information:
https://community.ieawind.org/blogs/cezanne-murphy-levesque/2018/09/17/iea-wind-tcp-2017-annual-report (wind energy developments in countries associated to IEA Wind, and a
description of several research tasks)
Contents
Part 1
•
Introduction, Costs and contribution to energy demand,Global situation Wind energy, Challenges
•
Wind ResourcesPart 2:
•
Energy Conversion•
Design loads•
Technology•
Wind farm effects•
Operation and Maintenance87
Wind speed, wind energy and wind power
V m
1 m2
A
ρ = 1.22 kg/m3
wind power density =
wind energy per second through unit area (A = 1m2)
Pw = ½ ρ V2 V
*
1 = ½ ρ V3 [W/m2]V [m/s]
V ρ Wind power density Pw
4 1.22 39 6 1.22 131 6 1.00 108 8 1.22 312 8 1.00 256 10 1.22 610 V3 makes site selection essential!!
88 5-2-2019
The wind as source of energy
0 2 4 6 8 10 12 14 16 18 3 5 7 9
Annual average wind speed at 50m height [m/s] G en er at io n c o st i n € ce n ts /k W h 40m hub height 55m hub height 75 hub height
Long term mean wind speed at h=10m
5-2-2019 89
Expressions for wind shear V(H)
a between ~ 0.1 (flat terrain) and 0.4 (hilly or urban terrain)
H V H Href = H Href
[
VH Vref]
a VH Vref = log (H/z0) log (Href/z0) z0 = 0.0002 m - 1mVertical gradient or ‘vertical wind shear’
5-2-2019 90
Typical values of roughness length z
0Roughness class 0 z0 ~ 0.0002
Roughness class 1 z0 ~ 0.03
Roughness class 2; z0 ~ 0.1 Roughness class 3; z0 ~ 0.4
(roughness classes as defined in the Danish norm, see also European Wind Atlas)
91
The (annual) mean power of the wind
•
The (annual) mean power depends on thefrequency distribution of the wind speeds and is larger than the instantaneous power at the
(annual) mean wind speed.
•
Example: mean wind speed = 4 m/sCase A: constant 4 m/s
4
time
Mean power: 39 W/m2
Case B: half of the time 0 m/s, other half 8 m/s 4 8 0 time Mean power: 156 W/m2
96
Frequency distribution by binning method
• Wind speed intervals of 0.5 m/s width, centred around integer and half integer values, e.g. from 6.75 to 7.25 m/s.
• Count the relative number of 10 minute means in the different bins Note many figures in the following
use an interval of 1 m/s which
however does not comply with the standards
97
Frequency distribution graph of wind
speed and energy density
A wind turbine is characterised by a power curve (i.e. electrical power as function of wind speed)
Vrated
Vcut_in Vcut_out
Prated: This represents the often mentioned ‘installed capacity’ (or ‘installed power’) of a wind turbine
The energy estimation
AEP(V=7) = P(V=7) * 0.05 * 8760 kWh/yr
• (Electrical) power produced by given turbine is function of
(stationary) wind speed: P-V curve • Energy is power*time
• If wind speed is in interval from (e.g.) 6.5 to 7.5 m/s for 5% of time (nr of hours/year = 8760) then contribution to annual energy production (AEP):
Estimation of annual energy production (AEP) or
output (AEO)
wind speed relative number of P(V) AEP(V) interval frequency hrs/yr kW kWh/yr
…… …x.. 8760*x …... …… 7.0 0.05 438 500 219000 ……. ……. …….. ….. …….. ……. ……. …….. ….. …….. ________ _______ _______ S = 1 S = 8760 S = AEP 100
101
Estimation of AEP
•
Previous slide is ‘clean’ estimate.•
For realistic estimate account for:- availability (95% (guaranteed, generally higher) - array losses (operation in wakes in wind farm)
(typical value 10%)
- electrical conduction losses (typical value 2-5 %)
Many people characterise a wind turbine by the CAPACITY FACTOR
Capacity (or plant) factor: CF = Annual energy production Nominal power (kW) * 8760 hr
Capacity factor = Annual Energy Production divided by the
energy production when the turbine operates continuously at
rated (or nominal) power
102
The value of CF should preferably not be less than 0.25
A good site can give a CF of 0.5 or higher !!
Wind turbine characteristics and the wind regime P2 P3 P1 % V [m/s] Jos Beurskens V [m/s] Vrated 103
Capacity factor in relation to wind turbine,
generator and wind climate
1 2 3 Vrated/Vave 0.4 0.2 0.1 C ap ac it y fa ct or 2 3 1 Vrated/Vave 104
105
Analytical wind speed frequency distributions,
the Weibull distribution
106
Wind speed distribution graphs
Measured distribution can usually be approximated by a ‘Weibull’ distribution
=
k kA
V
A
V
A
k
V
f
(
)
exp
1 % V [m/s] V [m/s] g g Vgem= 4 m/s Vgem= 6 m/s k shape factor A scale factor107
Analytical frequency distribution,
probability density function
• Measured wind speed frequency distribution and annual mean wind speed based on it can be used to characterize wind
resource, but much easier, especially for classification:
• Analytical frequency distribution with few parameters that gives good approximation of measured distributions:
• Based on: Cumulative distribution function F(V):
108
Weibull distribution I
Weibull distribution:
Two parameters: A (or c) [m/s] is scale factor
k [-] is shape factor (k=2, Rayleigh distribution)
F V
P V
V
V
A
k( )
(
)
=
exp
1
General properties of any cumulative distribution function F(0) = 0, wind speed never less than 0
109
Weibull distribution II
•
From the cumulative distribution F the frequencydistribution f is obtained by differentiation:
•
The relative frequency with which wind speed is ininterval ΔV around V is equal to f DV and also equal to:
F(V + ½ ΔV) - F(V - ½ ΔV)
F V
V
F V
V
V
dF
dV
f V
f
dF
dV
k
A
V
A
V
A
k k(
)
(
)
exp
=
=
=
=
1 2 12 1D
D
D
D
Hence
110
Comparison Weibull fit with measured
distribution
111 5-2-2019
• PMF (probability mass function) counts the relative frequency of 10 minute
means in the different bins, e.g. 6% of the time the wind speed is between 5.75 and 6.25 m/s
• Weibull PDF (probability distribution function) is a continuous function
- The relative frequency with which
wind speed is in interval ΔV around V can be found from it as f ΔV
- E.g. f(6) is 0.12
- The relative frequency with which the wind speed is between 5.75 and 6.25 m/s is
f(V) ΔV = 0.12 * 0.5 = 0.06
Mind the difference between a PMF (discrete) and PDF(continuous)
112 5-2-2019
Some observations on typical values of k and A
• It can be shown that Vmean ~0.9 A
• Usually higher k values go together with higher Vmean values
• Common k value for Northern European coastal regions at 10 m height a.g.l.: 1.9 to 2.0. Inland values smaller e.g. 1.4
• Common k values for trade wind regions at 10 m height a.g.l.: 3 to 3.5 or even 4
• Typically k value increases (Wieringa) with height a.g.l. (less variations at more distance from surface):