111,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
Covers all the key areas of wind resource assessment technologies from an engineer's perspective * Focuses on wind analysis for wind plant siting, design and analysis * Addresses all aspects from atmospheric boundary layer characteristics, to wind resource measurement systems, uncertainties in measurements, computations and analyses, to plant performance * Covers the basics of atmospheric science through to turbine siting, turbine responses, and to environmental impacts * Contents can be used for research purposes as well as a go-to reference guide, written from the perspective of a hands-on engineer * Topic is of ongoing major international interest for its economic and environmental benefits
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 579
Veröffentlichungsjahr: 2015
Ähnliche
Table of Contents
Cover
Title Page
Copyright
Preface
Introduction
Acknowledgments
About the Author
List of Symbols
Chapter 1: Introduction
1.1 Wind Resource Assessment as a Discipline
1.2 Micro-siting Briefing
1.3 Cascade of Wind Regime
1.4 Uncertainty of Wind Resource
1.5 Scope of the Book
References
Chapter 2: Concepts and Analytical Tools
2.1 Surface Roughness and Wind Profile
2.2 Speed-up Effect of Terrain
2.3 Shelter Effect of Obstacles
2.4 Summary
References
Chapter 3: Numerical Wind Flow Modelling
3.1 Modelling Concept Review
3.2 Linearised Numerical Flow Models
3.3 Mass-Consistent Models
3.4 CFD Models
3.5 Meso Scale NWP Models
3.6 Inherent Uncertainties in Wind Flow Modelling
3.7 Summary
References
Chapter 4: Wind Park Physics and Micro-siting
4.1 Wind Power Density
4.2 Wind Power Conversion
4.3 Wind Turbine Wake Effects
4.4 Wind Turbine Micro-siting
4.5 Summary
References
Chapter 5: Wind Statistics
5.1 Statistics Concepts Review
5.2 Wind Data Time Series
5.3 Mean Wind Speed of the Whole Time Series
5.4 Weibull Distribution
5.5 Estimating Weibull Parameters
5.6 Extreme Wind Statistics
5.7 Summary
References
Chapter 6: Measure–Correlate–Predict
6.1 Wind Data Correlation
6.2 Wind Data Regression and Prediction
6.3 MCP Methodology for Wind Energy
6.4 MCP Uncertainty
6.5 Sources of Reference Data
6.6 Summary
References
Chapter 7: Wind Park Production Estimate
7.1 Gross and Net AEP
7.2 AEP Uncertainty Analysis
7.3 Natural Variability of Wind
7.4 Uncertainty in Wind Measurement
7.5 Uncertainty in Wind Flow Modelling
7.6 A Case Study
7.7 Wind Resource Assessment Report
7.8 Summary
References
Chapter 8: Measuring the Wind
8.1 Representativeness of the Met Mast
8.2 Cup Anemometer Physics
8.3 Met Mast Installation
8.4 Met Mast Operation and Maintenance
8.5 Data Validation
8.6 Alternative Wind Sensors
8.7 Summary
References
Chapter 9: Atmospheric Circulation and Wind Systems
9.1 General Concepts
9.2 Laws and Driving Forces
9.3 General Atmospheric Circulations
9.4 Synoptic Scale Wind Systems
9.5 Meso-scale Wind Systems
9.6 Micro-scale Winds
Summary
References
Chapter 10: Boundary Layer Winds
10.1 Atmospheric Stability
10.2 Orographic Effects
10.3 Onshore Boundary Layer Winds
10.4 Offshore Boundary Layer Winds
10.5 Summary
References
Chapter 11: Environmental Impact Assessment
11.1 Biological Impacts
11.2 Visual Impacts
11.3 Noise Impacts
11.4 Weather and Climate Change
11.5 Public Health and Safety
11.6 Summary
References
Appendix I: Frequently Used Equations
Appendix II: IEC Classification of Wind Turbines
Appendix III: Climate Condition Survey for a Wind Farm
III.1 Calculating the Ambient Temperature Range
Appendix IV: Useful Websites and Database
Index
End User License Agreement
Pages
xiii
xiv
xv
xvii
xix
xxi
xxii
xxiii
xxiv
xxv
1
2
3
4
5
6
7
8
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
271
272
273
275
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
Guide
Cover
Table of Contents
Preface
Introduction
Begin Reading
List of Illustrations
Chapter 1: Introduction
Figure 1.1 Energy flux of the wind at 850 hPa (about 1500 m.a.s.l.) in W/m
2
from 8 years of the NCEP/NCAR reanalysis [2]
Figure 1.2 Wind power density distribution of China
Figure 1.3 The variation (slightly exaggerated) of mean wind speed 10 m.a.g.l. due to topographical effects (full line) for typical conditions of Denmark [5]
Figure 1.4 Reanalysis data for a point in Ireland showing the variability of the mean wind speed
Chapter 2: Concepts and Analytical Tools
Figure 2.1 Idealised wind profiles for three different surface types
Figure 2.2 Sketch of a wind profile change after passing through a coastline that separates two distinctive roughness surfaces, sea on the left side of the dashed line and land on the right. Arrows represent the wind direction
Figure 2.3 Idealised scenario of the wind profile before and after a roughness change
Figure 2.4 Illustration of wind profile blowing across a dense forest
Figure 2.5 Indication of the mean wind speeds at two heights
Figure 2.6 Streamlines of wind flow over an idealized hill
Figure 2.7 Perspective plot of Askervein hill and the layout of the masts [1]
Figure 2.8 Horizontal fractional speed-up ratios for wind flow over Askervein hill at 10 m above ground level. Hollow rectangles: wind tunnel data; solid rectangles: field measurement; solid lines: three-dimensional numerical model; dotted lines: two-dimensional numerical model; dashed lines: theoretical model [11]
Figure 2.9 Simultaneously recorded undisturbed upstream wind profile and the profile on top of the Askervein hill (Jensen
et al
., 1984 [13]). The symbols represent wind speed measurements. Both profiles correspond with those in Figure 2.10
Figure 2.10 Flow over an idealized hill with undisturbed upstream and hill-top wind profiles. Two characteristic length scales are also indicated:
L
, the half-width of the middle of the hill, and
l
, the height of maximum relative speed-up [1]
Figure 2.11 Wind flow over an escarpment blowing from the sea to land
Figure 2.12 Virtual surface created by flow separation indicated by vortexes. Arrows represent wind direction; dashed line the virtual surface
Figure 2.13 Cat. I: Simple flat [1]
Figure 2.14 Cat. II: Simple hilly [1]
Figure 2.15 Cat. III: Complex hilly [1]
Figure 2.16 Cat. IV: Extremely complex mountains [1]
Figure 2.17 RIX calculated by WAsP Map Editor. Line segments with slopes greater than a certain threshold are indicated by the thick radial lines [16]
Figure 2.18 Percentage reduction of wind speed due to shelter by a two-dimensional obstacle [1]
Figure 2.19 Sketch of the shelter zone by a three-dimensional obstacle
Figure 2.20 Side view (left) and top view (right) of wind flow around an obstacle, indicating more pronounced turbulent airflow downstream [20]
Figure 2.21 Example of an obstacle-like peak. The wind turbine in the rectangle may suffer from severe turbulence generated by the peak in the front
Chapter 3: Numerical Wind Flow Modelling
Figure 3.1 Example of a structured mesh (WAsP CFD) [16]
Figure 3.2 The European Wind Atlas methodology [28]
Figure 3.3 Grid structure of the zooming polar grid in WAsP [28]
Figure 3.4 WAsP wind speed prediction error as a function of Δ
RIX
. Square dots are the log errors of the 25 predicted wind speeds and the solid line is the fitted line [36]
Figure 3.5 Types of interactions considered in an NWP model
Figure 3.6 Demonstration of the butterfly effect
Chapter 4: Wind Park Physics and Micro-siting
Figure 4.1 Betz's tube indicating the wind flow changes before and after hitting the rotor (S)
Figure 4.2 Thrust coefficient
C
T
curve of typical pitch-regulated wind turbines
Figure 4.3 Power curve and
C
e
curve of a typical pitch-regulated wind turbine at air density of 1.225 kg/m
3
Figure 4.4 Wake of the Horn's Rev offshore wind park
Source
: Vattenfall, photograghed by Christian Steiness
Figure 4.5 Illustration of the wake structure of a horizontal-axis wind turbine
Source
: Flodesign Inc.
Figure 4.6 The structure of the Jensen wake model.
U
0
represents the free-stream velocity,
U
the velocity in the wake at the distance of
X
behind the rotor disc with a diameter of
D
and
D
W
is the diameter of wake at distance
X
[12]
. Source
: Risø National Laboratory
Figure 4.7 Turbines taken into account in the Frandsen model. Solid circles with an arrow pointing to the wind turbine in consideration are taken into account, while the crossed circles are not [26]
Source
: Risø National Laboratory
Figure 4.8 Illustration of the development of the internal boundary layer (IBL) in a deep-array wind farm, represented by the two curves
Figure 4.9 Wind turbine layout with a regular grid
Figure 4.10 Wind frequency rose (left) and energy rose (right) for the same set of wind data created with WindPRO 2.7
Figure 4.11 Typical micro-siting procedure
Figure 4.12 Site survey for a wind park in Southern China
Figure 4.13 Wind sector management
Chapter 5: Wind Statistics
Figure 5.1 Transient wind speed data points recorded every 2 seconds for 10 minutes (600 s). Dashed line in the centre represents the mean wind speed value over this period
Figure 5.2 Generalised turbulence intensity and mean wind speed curve
Figure 5.3 Illustration of trended transient wind speed samples. Solid line represents the linear trend of the sample data; dashed line in the centre is the mean wind speed value over this period
Figure 5.4 Standard deviations of different wind speeds and averaging times (from the top: 30 min, 10 min, 5 min and 1 min), all normalised relative to a 60-minute averaging time for neutral stability
Figure 5.5 Random wind direction measurements. Same mean direction (zero degree) with different spreads of the data points
Figure 5.6 Two-parameter Weibull probability density function
Figure 5.7 Weibull PDF fitted with wind data measurements for all direction sectors
Figure 5.8 Weibull CDF fitted with wind data measurements (same data as in Figure 5.7) of all direction sectors
Figure 5.9 Gumbel regression of extreme wind speeds
Figure 5.10
Chapter 6: Measure–Correlate–Predict
Figure 6.1 Examples of correlation coefficient plotted against the averaging interval. The four data points on each line represent averaging intervals of 6 hours, 1 day, 7 days and 30 days respectively
Figure 6.2 Wind power density (WPD) of three wind data series of distinctive wind climates (Shandong, Inner Mongolia and Yunnan, China) plotted against the averaging interval. The data points on each line represent averaging intervals of 1, 2, 3, 6 and 12 hours respectively
Figure 6.3 WPD of the same wind data series as Figure 6.2 plotted against the averaging interval of wider ranges (left: 1 h to 20 h; right: 10 days to 30 days)
Figure 6.4 Wind speed time series of target and reference site
Figure 6.5 Wind speed prediction (interpolation) using the simple linear regression equation
Figure 6.6 Wind speed prediction (extrapolation) using the simple linear regression equation
Figure 6.7 Simple linear regression of concurrent wind speeds in Figure 6.4
Figure 6.8 Sample data and first-order model for the wind speed-up (
x
= wind at reference and
y
= at speed-up) [15]
Source
: EMD International A/S
Figure 6.9 Sample data and first-order model for the wind speed-up (
x
= wind at reference and
y
= wind veer) [15]
Source
: EMD International A/S
Figure 6.10 Illustration of a single neuron (left) and the layered structure of a feedforward ANN (right)
Figure 6.11 Measured normalised standard deviation (standard deviation/mean wind speed) of the mean wind speed for six inland and six offshore pairs of data [26]
Figure 6.12 Dependence of the standard error of a 10-year predicted mean speed on the number of years of historical data used in the prediction, produced by sampling reanalysis data at every grid point in the United States. The result for a sample of rawinsonde stations is shown for comparison [30]
Source
: AWS Truepower, LLC
Chapter 7: Wind Park Production Estimate
Figure 7.1 Probability density curve of a normal distribution, where
μ
and
σ
denote the mean and standard deviation respectively
Figure 7.2 Combination of independent uncertainties
Figure 7.3 P90 (shaded area) in normal distribution
Figure 7.4 An example curve of expected AEP against probabilities of exceedance in a normal distribution
Chapter 8: Measuring the Wind
Figure 8.1 Wind condition variations on Chuandao Island in Southern China. Wind data were all measured at 70 m above ground level for a full year
Figure 8.2 A cup anemometer
Figure 8.3 The actual wind speed
S
and the horizontal wind speed
U
, where
u
,
v
and
w
are the three directional components of the actual wind speed
S
in an orthogonal coordinate system
Figure 8.4 Percentage difference between the indicated wind speed and the true total wind speed for cup anemometer ‘A’ in the wind tunnel for various tilt angles at various wind speeds [2]
Figure 8.7 Percentage difference between indicated wind speed and true total wind speed for cup anemometer ‘B’ in the free atmosphere for various tilt angles at various wind speeds [2]
Figure 8.8 Schematic of a ‘well-designed’ cup anemometer [2]
Figure 8.9 Schematic of a poorly designed cup anemometer [2]
Figure 8.10 Iso-speed plot of the flow around a solid tubular tower. The local wind speed is normalised by the free-field wind speed attacking left to right [7]
Figure 8.11 Iso-speed plot of the flow around a triangular lattice tower. The local wind speed is normalised by the free-field wind speed attacking from left to right [7]
Figure 8.12 A wind direction vane
Figure 8.13 Example of good practice (left) and poor practice (right) in mounting a cup anemometer on the side of a met mast
Figure 8.14 Panorama view at the met mast location, to the north
Figure 8.17 Panorama view at the met mast location, to the west
Figure 8.18 A vertical propeller anemometer
Figure 8.19 A horizontal propeller anemometer with a single tail vane
Figure 8.20 A two-axis sonic anemometer
Figure 8.21 A sodar system
Figure 8.22 A lidar system
Chapter 9: Atmospheric Circulation and Wind Systems
Figure 9.1 Vertical structure and temperature profile of the atmosphere
Figure 9.2 Kinetic energy spectrum near the surface in mid-latitudes [2]
Figure 9.3 Spatial and time scales of typical meso- and micro-scale weather processes [3]
Figure 9.4 The process of pressure gradient force and Coriolis force reaching equilibrium on a constant height surface in free atmosphere aloft in the northern hemisphere
Figure 9.5 The vertical variation of geostrophic wind in a barotropic atmosphere (a) and in a baroclinic atmosphere (b). The dotted lines enclose isobaric surfaces, which remain at a constant slope with increasing height in (a) and increase in slope with height in (b)
Figure 9.6 Three cell circulation in the northern hemisphere
Figure 9.7 Polar front snapshots of 26 May 2006 (left) and 30 May 2006 (right) in the northern hemisphere
Figure 9.8 Flow structures of a cyclone (left) and an anticyclone (right)
Figure 9.9 The structure of a typical tropical storm
Figure 9.10 An illustration of a fully developed sea breeze circulation
Figure 9.11 An illustration of fully developed mountain–valley and mountain–plain winds at daytime (bottom) and at night-time (top)
Figure 9.12 Illustration of katabatic winds
Figure 9.13 Main turbulence generation mechanisms in the atmospheric boundary layer
Figure 9.14 An illustration of the surface energy balance over land
Chapter 10: Boundary Layer Winds
Figure 10.1 Three cases of vertical parcel displacement in neutral (left), unstable (centre) and stable (right) surrounding atmosphere.
T
denots temperature;
z
denotes height
Figure 10.2 The influence of atmospheric stability on a wind profile in the surface layer. Dashed line: neutral; solid line: stable; dot-and-dashed line: unstable
Figure 10.3 Sketch of forced channelling of the wind
Figure 10.4 Sketch of pressure-driven channelling of the wind
Figure 10.5 Unstable (left) and relatively stable air flow over (centre) and around (right) a mountain of various heights
Figure 10.6 Winds flow over a mountain in three different prone-stable conditions. PE represents potential energy and KE represents kinetic energy
Figure 10.7 NASA satellite image (Landsat 7) of clouds off the Chilean coast near the Juan Fernandez Islands on 15 September 1999
Figure 10.8 NASA satellite image (MODIS imager on board the Terra satellite) of a wave cloud forming off Amsterdam Island in the far southern Indian Ocean. Image taken on 19 December 2005
Figure 10.9 Vertical structure of the atmospheric boundary layer over flat and homogeneous terrain
Figure 10.10 Dirunal cycle of boundary layer structure in clear sky conditions over land [1]
Source
: Springer Science+Business Media (adapted)
Figure 10.11 Diurnal variations in mean wind speeds measured at different heights, plotted from a year of wind data from a met mast located in Northest China
Figure 10.12 An example of an observed wind profile in the presence of a nocturnal low-level jet
Figure 10.13 Sketch of developing IBL over a step-change in surface property
Chapter 11: Environmental Impact Assessment
Figure 11.1 Sketch of blade shadow flicker on a residential building
Figure 11.2 V90-2MW noise curves, corresponding to Table 11.1
Figure 11.3 Example of a turbine noise propagation map, created with WindPRO [23]
Source
: EMD International A/S
Figure 11.4 Addition of two sound levels
Figure 11.5 Estimated noise levels at different distances from a typical 2 MW wind turbine. The sound power level at the turbine is 104.0 dB(A)
Appendix III: Climate Condition Survey for a Wind Farm
Figure b03.1 Example page from weatherbase.com
List of Tables
Chapter 2: Concepts and Analytical Tools
Table 2.1 Roughness length and class for typical surface characteristics [1]
Chapter 4: Wind Park Physics and Micro-siting
Table 4.1 Wind turbine classification derived from IEC 61400-1 (2005) for the standard air density of 1.225 kg/m
3
[37]
Chapter 5: Wind Statistics
Table 5.1 Example of mean wind speed calculations
Table 5.2
Table 5.3 GEV types and their properties [27]
Table 5.4
Chapter 6: Measure–Correlate–Predict
Table 6.1 Quality of reference data
Chapter 7: Wind Park Production Estimate
Table 7.1 Production loss due to turbine performance [13]
Table 7.2 Overall uncertainty reduced after taking 1% off each component
Table 7.3 Main sources of uncertainty in wind measurement (cup anemometer)
Table 7.4 Uncertainty of the vertical extrapolation modelled by a logarithmic profile
Table 7.5 Uncertainty of horizontal extrapolation due to the average distance to the mast
Table 7.6 Uncertainty of horizontal extrapolation due to overall terrain complexity
Table 7.7 Uncertainty of horizontal extrapolation in wind flow modelling for a distance of every 1 km from the mast to the turbine in question
Table 7.8 Uncertainty of wind flow modelling due to wind resource similarity
Table 7.9 Uncertainty components of the case study
Table 7.10 P90 and P75 AEPs of the wind farm in the case study (GW h/y)
Table 7.11 Percentage uncertainties in AEP associated with the two masts
Chapter 8: Measuring the Wind
Table 8.1 Good practice for met mast installation
Table 8.2 Example of a met mast description
Table 8.3 Example of a data logger system description
Table 8.4 Example of the equipment list
Table 8.5 Checklist of an on-site inspection
Table 8.6 Example of a met mast monitoring log
Chapter 9: Atmospheric Circulation and Wind Systems
Table 9.1 Standard atmosphere at sea level
Chapter 11: Environmental Impact Assessment
Table 11.1 V90-2MW noise curve mode 0 (dB) at 1.225 kg/m
3
Table 11.2 Sound levels associated with everyday sounds and wind turbines in dB(A)
Wind Resource Assessment And Micro-Siting
Science And Engineering
Matthew Huaiquan Zhang
This edition first published 2015
© 2015 China Machine Press. All rights reserved.
Published by John Wiley & Sons Singapore Pte. Ltd., 1 Fusionopolis Walk, #07-01 Solaris South Tower, Singapore 138628, under exclusive license granted by China Machine Press for all media and languages excluding Simplified and Traditional Chinese and throughout the world excluding Mainland China, and with non-exclusive license for electronic versions in Mainland China.
For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com.
All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as expressly permitted by law, without either the prior written permission of the Publisher, or authorization through payment of the appropriate photocopy fee to the Copyright Clearance Center. Requests for permission should be addressed to the Publisher, John Wiley & Sons Singapore Pte. Ltd., 1 Fusionopolis Walk, #07-01 Solaris South Tower, Singapore 138628, tel: 65-66438000, fax: 65-66438008, email: [email protected].
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data applied for.
Preface
I joined Vestas China in early 2008 when China was undergoing a tremendous boom in the wind energy industry with the installed capacity of wind power more than doubling year on year. It was indeed an exciting time, but very soon I realised that the industry was literally running with blind eyes. Wind data were measured carelessly. Projects were excecuted with merely a few months' poor wind data and being built where power grids were already saturated with wind energy. The number of wind turbine manufacturers went from a few to close to a hundred in just a few years! Wind resource assessment and turbine siting was considered unimportant and had to very often give way to the top-to-bottom bureaucratic planning of a speedy development. I have therefore felt the urge to bring sense back to the industry ever since I started in the industry. At the time, there was no book in the Chinese market on this subject and the wind engineers were generally equipped with insufficient knowledge to deliver a sound wind resource assessment, owing at least partly to the low requirements from the industry. I then decided to write one myself. After a year of hard work, the first edition of the book was finally published by China Machine Press in June 2013. The market has responded very positively.
I have to admit that writing a technical book while keeping a full-time job is quite tough. It basically means no weekends, no holidays and working until midnight nearly every day for a whole year. It was all worth it in the end but I certainly wished that I would never put myself into the same kind of stress ever again. Yet, as soon as the Chinese edition was published, I pretty much had to immediately come to terms with writing it in English all over again. It all happened because my partner and I made the decision of relocating to the UK, but doing so meant leaving 10 years of life in Beijing behind. I therefore got in contact with John Wiley & Sons, Ltd in the autumn of 2013 in order to pursue the publication of the book in English, and managed to come to an agreement with them. As a second-language speaker who only learnt the language at school, writing it in English seemed overly ambitious. This time, though, I resigned from my full-time job and became an independent consultant so that I could gain a better control of my time as well as working on the process of my relocation to the United Kingdom. Instead of directly translating the content into English, I have taken the opportunity to refine the areas where I did not have time to do so in the Chinese edition and also catering to an international audience. Consequently, the book you are reading now is essentially the second edition with sizable improvements and reshuffling of the first one.
My educational background, an MSc in Material Science and second major in International Economy, has little to do with the domain of wind energy application. Also, before joining Vestas, I worked on website development. It may seem a bit sporadic but it also makes the book interesting because it is composed by a hands-on specialist who first approached the subject as a total stranger and gained a high level of proficiency. The structure and content of the book are carefully selected and designed on the basis of my own learning experience of the subject. The domain of wind resource assessment and micro-siting involves multiple disciplines ranging from statistics and flow modelling to atmospheric physics and meteorology. Each chapter in this book can be a much specialised area of study, easily filling its own books. This may be one of the reasons why this subject has rarely been organised into one structured and coherent text. Therefore, the content in this book does not intend to be exhaustive or overly precise. Many aspects of wind resource assessment are still developing and undergoing sometimes heated debates, such as numerical wind flow modelling and uncertainty analysis. I, myself, am still learning new things and gaining more experience every day. The aim of this book is therefore not to decisively settle every debate, but I do hope that it can provide the readers with sufficient information and insight to make sound decisions in each critical step of wind resource assessment and in delivering more wind power projects that perform effectively. For those who wish to further their knowledge in any specific topics or enter academics, this book can still serve as a gateway or a good foundation.
Matthew Huaiquan Zhang
Introduction
The successful development of wind energy projects depends on an accurate assessment of the wind conditions and siting of each wind turbine. Wind in nature is a very complex meteorological phenomenon with a dominant feature of constant and sometimes violent fluctuations. From measured wind data, which often only covers a time range of a couple of years, to the estimate of 20 years of average wind power production of the wind farm, it takes multiple complicated analyses with inherited risks. A mistake in this stage of evaluation can cause severe financial losses and missed opportunities for developers, lenders, and investors.
Wind Resource Assessment and Micro-siting: Science and Engineering constructively and coherently pulls together all the key theories of the domain, aiming to form a strong, systematic foundation of wind resource assessment and micro-siting for readers. It should allow readers to utilise the contents for research purposes as well as a go-to guide for useful information. The areas covered in this book include analytical and numerial wind flow modelling, wind statistics, wind measurement techniques and data analysis, MCP, uncertainty analysis, wind energy meteorology, offshore micro-siting and environmental impact assessment. In order to assist readers' learning, most topics start with easy-to-understand background details and ease into the practical demonstration of day-to-day engineering work. The author brings his own experience to bear on the teaching and applications of his knowledge acquired over the years, which makes it reader friendly. It aims to build a bridge between general professionals, through to more advanced and specialist researchers in each topic.
Acknowledgments
I would like to thank Vestas Wind System A/S, the organisation that brought me into the domain of wind resource assessment and provided me with extensive training from the very beginning. A few topics are inspired by and built upon what I learnt in the training programmes, for example wind data analysis and uncertainty analysis delivered by Kim E. Andersen and meso-scale meteorology by Line Gilstad. I am also grateful for the permissions to use and reproduce many of the figures and contents in the book by organisations including Risø National Laboratory, DTU, IEA Wind R&D and EMD International A/S, and the support from both China Machine Press and John Wiley & Sons, Ltd. Notwithstanding their diligent efforts and support, any errors and oversights remain the sole responsibility of the author.
Matthew Huaiquan Zhang
About the Author
Receiving his Masters degree in 2006 from Tsinghua University, China, the author entered the wind energy industry in 2008. He began his career as a wind and site engineer with Vestas, the world's leading wind turbine manufacturer, gaining WAsP certification in 2009 in a training and examination programme in Chicago, USA. Matthew Huaiquan Zhang has dedicated himself to gaining expert knowledge in all aspects of wind resource related technologies. Over the years, he has successfully assessed over 5 GW of wind park capacity and trained a team of competent engineers for the Chinese wind energy market.
List of Symbols
a
scale parameter in Weibull and generalised Pareto distribution
fractional wind speed reduction along the rotor surface normal
slope parameter of a straight line
acceleration
b
scale parameter of a straight line
b
j
threshold at neuron
j
(ANNs)
d
displacement height
e
residual in regression models
f
frequency
nomalised frequency in Equation (8.18)
Coriolis parameter
f
(
x
)
probability density function
regression function
g
Earth's gravity
h
height
h
H
wind turbine hub height
k
shape parameter in Weibull, extreme value and Pareto distribution
entrainment constant in Jensen wake model
kurtosis
factor in Equation (
5.89
)
l
height where the maximum speed-up effect is found
m
mass
n
number of records
frequency in Equation (8.18)
p
number of fitted parameters in Equation (6.8)
q
specific humidity (mixing ratio) of the air mass
r
radius
s
c
mean crosswind relative spacings in a regular grid layout
s
d
mean downwind relative spacings in a regular grid layout
s
i
relative distance to the neighbouring turbine
i
t
time (scale)
u
horizontal wind speed
turbulence fluctuation component in horizontal direction
u′
turbulence term of horizontal component
u
*
friction velocity
u
0
horizontal wind speed at the top of the hill
v
wind speed
perpendicular wind speed
turbulence fluctuation component in perpendicular direction
v′
turbulence term of perpendicular
v
i
i
th wind speed record in a wind data series
w
turbulence fluctuation component in vertical direction
vertical wind speed
w′
turbulence term of vertical component
w
j
weight of
j
th data point
x
distance to the roughness change line
horizontal axis of an orthogonal coordinate system
a random variable
y
perpendicular axis of an orthogonal coordinate system
z
height level above surface
vertical coordinates
z
r
reference height level
z
0
roughness length
z
00
wind-farm equivalent roughness length
AEP
annual energy production
A
W
section area of turbine wake
location A in WAsP model
B
location B in WAsP model
Ce
electrical power coefficient
CF
capacity factor
C
p
power coefficient
C
T
thrust coefficient
D
rotor diameter
D
eff
effective rotor diameter
D
i
Cook's distance
D
W
diameter of wind turbine wake
E
operator of mean
E
PF
energy pattern factor in Equation (
5.60
)
Err
prediction error
F
force
F
c
centrifugal force
F
g
gravitational force
F
f
friction force
F
j
cumulative probability of ranked data point
i
(Gumbel method)
F
r
Coriolis force
Froude number
F
p
pressure gradient force
F
T
cumulative probability of a
T
year period
F
(
x
)
cumulative distribution function
G
gust factor
I
turbulence intensity
I
0
ambient turbulence intensity
I
eff
effective turbulence intensity
I
ref
average turbulence intensity at wind speed of 15 m/s
I
u
streamwise (horizontal) turbulence intensity
I
T
maximum turbulence intensity in the centre of turbine wake
I
W
wake added turbulence intensity
K
kinetic energy
K
m
turbulent transfer coefficient
L
half-width of the middle of the hill
loss in annual energy production
Monin–Obukhov length
L
p
sound pressure level (dB) at the receptor
L
w
sound power level from a noise source (dB)
LAE
least absolute error
MSE
mean square error
N
number of neighbouring wind turbines
number of records
static atmospheric stability parameter
N
M
number of full years of onsite wind measurement
N
P
financial horizon of a wind power project
N
ref
number of years of the reference dataset (MCP)
N
target
number of years of the concurrent dataset at target site (MCP)
ORO
effects of orography in WAsP model
OBS
effects of obstacle in WAsP model
P
air pressure
wind power density
probability
sound power intensity
P
e
electrical power of a wind turbine generator
P
free
energy production at free-stream wind speeds
P
park
predicted energy production of a wind park (wake effect included)
P
r
air pressure at a reference level
z
r
P
w
wake probability
P95
annual energy production with 95% probability of exceedance
P90
annual energy production with 90% probability of exceedance
P75
annual energy production with 75% probability of exceedance
P50
annual energy production with 50% probability of exceedance
R
specific gas constant for air
distance from a sound source
R
2
coefficient of determination
correlation coefficient
Re
Reynolds number
RIX
ruggedness index
RMSE
root mean square error
ROU
effects of roughness (WAsP model)
S
sensitivity (production uncertainty)
upstream cross-section of roughness element
A
section area
turbine swept area/rotor area
S
(
n
)
spectral power density
SS
err
sum of squares of error/residual
SS
tot
total sum of squares
SSR
sum of squared errors
T
recurrence period
temperature
T
H
highest monthly-average temperature
T
L
lowest monthly-average temperature
T
v
virtual temperature
Tr
air temperature at a reference level
z
r
TKE
turbulence kinetic energy
U
mean wind vector component in horizontal direction
U
0
free stream horizontal mean wind speed
U
g
geographic wind vector
U
H
free stream hub height wind speed
wake-affected hub height wind speed
Up
predicted wind speed
U
M
measured wind speed
V
mean wind vector in perpendicular direction Volumn
V
50
50-year extreme mean wind speed
V
ave
annual mean wind speed at hub height
V
gust
gust wind speed
V50,
gust
50-year extreme gust wind speed
median value of wind speed bin
i
V
ref
reference wind speed in IEC standard
V
T
extreme wind speed of recurrence period
T
W
mean wind vector in vertical direction
W
reg
common regional wind climate (WAsP model)
W
A
micro-scale wind climate of location A (WAsP model)
W
B
micro-scale wind climate of location B (WAsP model)
X
a variable
distance
mean term of any meteorological variable
X′
fluctuation term of any meteorological variable
predicted value of
Y
(MCP)
α
wind shear component
constant in Equation (4.40)
scale parameter in generalised extreme value distribution
Charnock parameter
frequency-dependent sound absorption coefficient
β
initial wake expansion parameter
location parameter in generalised extreme value distribution
γ
skewness
λ
crossing rate in Equation (5.88)
σ
standard deviation/uncertainty
σ
1
intrinsic uncertainty of the 1-year mean wind speed
σ
ann
standard deviation of the annual mean wind speeds
σ
conbined
combined uncertainty
σ
i
uncertainty component
σ
H
standard deviation of average high tempeture
σ
L
standard deviation of average low tempeture
σ
N
intrinsic uncertainty of the
N
-year mean wind speed
σ
P
uncertainty in long-term windiness
σ
u
standard deviation of streamwise (horizontal) velocity
σ
v
sample standard deviation of wind speed
σ
v,hh
uncertainty of wind speed at hub height due to vertical extrapolation
σ
V
standard deviation of mean wind speed
θ
potential temperature
θ
i
i
th wind direction record in radius
i
th wind direction record in degree
θ
v
virtual potential temperature
τ
w
surface shear stress
ρ
air density
μ
mean of a random variable
μ
n
n
th central moment
μ
v
sample mean wind speed
μ
x
easting mean of directional components
μ
y
northing mean of directional components
μ
θ
sample mean wind direction in degrees
ρ
i
i
th air density record in a wind data series
κ
Kármán constant
turbulent kinetic energy (TKE)
ϵ
rate of dissipation of turbulence kinetic energy
spread of wind direction data points
η
e
electrical efficiency of a wind turbine generator
η
m
mechanical efficiency of a wind turbine generator
η
Park
wind park efficiency
ξ
location parameter in generalised Pareto distribution
ζ
stabliliy parameter
ω
angular velocity of the earth's rotation
ω
ij
weight on the input from neuron
i
to
j
(ANNs)
ϕ
latitude
Ψ
m
stability function
Γ
gamma function
Δ
α
uncertainty in wind shear
Δ
u
relative horizontal fractional speed-up (flow over hills)
ΔRIX
RIX difference between the predicted and reference sites
differential pressure gradient
Chapter 1Introduction
Energy supply is undoubtedly one of the most challenging issues facing human beings in the 21st century. Limited traditional fossil fuels are being used up gradually, let alone air pollutants and global warming caused by the combustion of those dirty fuels. Renewable energy has therefore attracted increasingly more attention in recent years. Wind energy, being one of the most commercially viable forms of renewable energy at the moment, has already played an important role in quenching our society's energy thirst. Yet there is still a long way to go before we can fully exploit the wind potential of our planet.
Wind is the ‘fuel’ for wind power generation. Its characteristics, that is wind conditions, are therefore of the upmost importance when it comes to determine the economics of a wind farm project. Wind conditions are set by nature, but how well we can understand or estimate them is another question. Wind resource assessment in essence is the estimation of wind conditions based on wind data available and topographical (roughness, obstacles and terrain) and meteorological (e.g. atmospheric stability, boundary layer structure, weather system) features of a given site.
Being invisible already makes it hard for us to picture the wind, and to make matters worse it varies constantly and dramatically in time and space, influenced by a great number of factors, some of which we may not even know of. However, on the other hand, building wind farms is very capital intensive and those wind farms have to generate profit for their owners. Profitability has to be predicted before wind farms are built with a reasonable risk premium. Such stringent requirements from the industry have raised sometimes almost impossible challenges for wind resource assessment professionals. After all, the results of wind resource assessment and micro-siting will determine the success of the investment of a wind power project.
This book endeavours to bring together pieces of core knowledge used in wind resource assessment and to put them into a logical order and to explain them, adding in the author's own experience obtained in day-to-day work scenarios. This kind of effort has rarely been made before, at least to the author's knowledge, even though a few publications covering a few sections of the domain can be found in the market.
1.1 Wind Resource Assessment as a Discipline
From a meteorological point of view, the study of a wind resource for the purpose of energy production can be described as wind energy meteorology, which has developed into an independent division of meteorology. In fact, a monograph named Wind Energy Meteorology by Emeis [1] has recently been published in early 2013, a milestone of the discipline. Petersen et al. [2] describe wind energy meteorology as applied geophysical and fluid dynamics, a combination of meteorology and applied climatology.
Despite its importance, wind energy meteorology has not been a major area of expertise required by the industry to produce satisfactory wind resource analysis results until the last decade or so. In the last decade especially, wind turbines have substantially grown in size and height, which means that they are exposed to much more complicated atmospheric boundary layer structures. Simplified engineering models, which worked well before, have to be re-examined based on the study of wind energy meteorology. The fact that wind turbines are usually erected in more complex terrain conditions, and even offshore nowadays, has also promoted the development of the discipline. Therefore a significant portion of time will be spent on this subject in order to form a physical profile of wind resource analysis for readers.
Wind resource assessment takes us one step closer to the wind energy industry, setting off from the ivory tower of physics. The domain of wind resource assessment should at least consist of wind data analysis, site analysis, wind turbine selection, wind turbine siting (micro-siting), wind flow modelling, power production estimates, wind park optimization and uncertainly analysis. Statistical tools are predominantly used in the process owing to the stochastic nature of the wind. Therefore, statistics becomes another pillar of wind resource analysis, the first one being the physical models explained by wind energy meteorology, such as the boundary layer profile and atmospheric stability. In order to ensure quality calculations, we need to understand how the wind should be measured as well as interaction mechanisms between wind and wind turbines and amongst turbines (wake effects).
The development of wind resource assessment has been accompanied and motivated by the commercial evolution of wind turbines and the construction of large-scale wind power projects. It will continue to do so in the foreseeable future. As a matter of fact, the expertise in wind resource assessment has become a core competence for many organizations in the industry and therefore well sought after.
1.2 Micro-siting Briefing
Micro-siting is really a meteorological definition, because in the eyes of a meteorologist, a few hundred metres is really on a micro scale. Micro-siting can be defined as the process of strategically positioning wind turbines within a given project area, in order to maximise power production with minimised turbine loads, that is optimising the wind park. Petersen et al. give an alternative definition of micro-siting, that is an estimation of the mean power produced by a specific wind turbine at one or more specific locations [2].
A full siting procedure includes considerations such as the availability and capacity of the power grid, the present and future land use, and so on, but these aspects are not considered in this book. However, one important issue concerning the siting of wind turbines is their environmental and health impact, such as noise and flickering, which can turn into a dominant factor in some cases and is explicated in Chapter 11.
1.3 Cascade of Wind Regime
The wind in nature almost never travels along a straight line; rather its track resembles circles. Those ‘circles’ are of all sizes, driven or dominated by different forces and induced by various mechanisms. Bigger ‘circles’ break into smaller ones and then even smaller ones until dissipated into heat, that is vibration of air molecules.
The wind we feel is a superposition of all the ‘circles’ of air movement at one spot. The scale of wind regime (or the size of the ‘circles’) can be described by two dimensions: temporal and spatial. The temporal scale and spatial scale of a wind regime are closely related. We can imagine that the bigger it is in space, the longer it takes to finish a circle. This cascade of wind regime should be the first physical model of the wind one should formulate before getting into the world of wind resource assessment. Wind regime is also referred to as wind climate or wind system. Chapter 9 will present wind systems of various scales in detail.
1.3.1 Global Scale Wind Regime
The atmosphere is a very complex heat engine whose energy is supplied by the heating of the earth's surface by the sun. Because the earth is tilted and also because of its uneven surface, different parts of the earth receive substantially different amounts of energy from the sun, which in turn induces air circulations with a spatial scale of the entire globe and a temporal scale of one or many years. This partly explains why wind resources are distributed so unevenly around the globe, as shown in Figure 1.1 [2].
Figure 1.1 Energy flux of the wind at 850 hPa (about 1500 m.a.s.l.) in W/m2 from 8 years of the NCEP/NCAR reanalysis [2]
Source: Risø National Laboratory
Long-term wind data measured around the globe are required to analyse wind climate on this scale, but such efforts are commonly hindered by poor data quality (usually measured at 10 m height and contaminated by local features and inconsistent through time) and insufficient measurement points.
In recent years, however, the advances in computational power, the availability of nontraditional meteorological datasets with global coverage (such as satellite data), in addition to the traditional ones used in the global meteorological network (e.g. the Global Observing System [3]), and the advances in weather prediction models have together made it possible to reconstruct the global scale weather situation at every instant over recent decades. Global meteorological models are able to provide dynamic, consistent wind data and statistics, while avoiding some of the setbacks associated with the direct use of wind data (in fact most reanalyses do not consider low-level wind data in the analysis because of their ‘contamination’ with local influences) [4]. Figure 1.1 [2] is a good example of such applications and indicates the global wind resource variation, though the figure is rather dated. Figure 1.2 demonstrates the distribution of wind power density in China.
Figure 1.2 Wind power density distribution of China
Source: China Meteorological Administration (CMA)
Global meteorological models are usually made with spatial resolutions too coarse to be used in project-based wind resource analysis. For a higher level of spatial resolution, which includes smaller-scale phenomena of significant influence on the wind resource, it is now a common practice to use mesoscale NWP models of much higher resolutions than global reanalysis models. These models cover a smaller domain using global data as boundary conditions. From the aspect of wind resource assessment and micro-siting for wind power projects, this global scale wind climate is a rather remote application, and therefore is generally excluded in this book. However, it does indicate the importance of long-term correction of in-situ wind measurements, which typically cover only a period of one or two years.
1.3.2 Synoptic Scale Wind Regime
We are probably more familiar with the synoptic scale wind regime because of the application of weather forecasts, which frequently mention weather systems, such as fronts, pressure systems, cyclones and anticyclones. Synoptic scale wind regimes usually have a spatial scale of more than 200 km (up to thousands of kilometres) and temporal scales of over a few days (up to a few months). Large synoptic scale wind systems (e.g. westerly winds in mid-latitude, trade winds and monsoons) are in most cases the main drivers of a local wind resource, although often severely altered by smaller scale phenomena.
Synoptic scale systems are obviously also too large for the purpose of analysing wind resource conditions within a wind farm and wind turbine siting in spite of its significant impact on local winds. The analysis of synoptic scale systems for wind turbine micro-siting only becomes relatively more important in some offshore cases where the overall atmospheric stability (one of the determining factors for wind turbine spacing and layout design) is closely related to wind direction and the passage of weather systems. Even then, it is usually not studied in detail. This is not to undermine the importance of understanding synoptic scale wind systems for wind resource assessment and micro-siting; rather it is to find the right focus for professionals working in this area as well as the scope of this book.
1.3.3 Meso-scale Wind Regime
Large scale models are insufficient to capture all weather phenomena related to the local wind resource. Meso-scale models zoom in from such large scale models and look at the missing scales between large scales and local or micro-scales. It covers special wind phenomena such as land and sea breezes, tropical hurricanes and convective storms. In the meantime, the meso-scale is also commonly used to explicit large scale flows modified by terrain features such as hills, mountains and surface features, which are not resolved in the larger scale models. As large scale and global models become more and more sophisticated and of increasing resolution, the division between large scale and meso-scale becomes blurred.
Meso-scale NWP models are commonly used nowadays in wind resource assessments in terms of a site hunt, that is searching for possible sites for wind power projects. In contrast with micro-siting, which deals with the wind resource within a given project area, a site hunt using meso-scale models can be referred to as macro-siting. Meso-scale NWP models are also able to generate a long-term wind data series, which is invaluable for long-term correlation of on-site wind measurements. As a result, a wind resource analyst should know by heart the correct usage of the results from meso-scale models and their pros and cons (see Chapter 3).
1.3.4 Local Scale Wind Regime
The local wind regime reflects local wind effects down to the smallest scales superimposing on larger scale features and meteorological systems. From a meteorological point of view, it generally represents micro-scale meteorological features and wind effects, giving the origin of the term ‘micro-siting’.
Local winds are driven by local geographical features. Near the surface, medium to small hills usually have similar levels of impact on wind speed with vegetation and obstacles [5]. Figure 1.3 illustrates the slightly exaggerated variation of mean wind speed at 10 metres above ground level on a typical Danish site, indicating clearly the effects of topographical features on local winds. During micro-siting, various micro-scale topographical features have to be carefully taken into consideration.
Figure 1.3 The variation (slightly exaggerated) of mean wind speed 10 m.a.g.l. due to topographical effects (full line) for typical conditions of Denmark [5]
Source: Risø National Laboratory
In general, in situ wind data measured by local meteorological masts (often referred to as a met mast in short) should be sufficient for the estimation of a wind resource at the measured point. However, it is more likely that we do not have measurement data at the specific point(s) of interest (at the hub height of the intended wind turbine at their precise locations) or not over a sufficiently long time (years) to be able to directly estimate the wind resource. Therefore, it is essential to have an understanding of the most important mechanisms that influence the wind locally. Micro-scale differs from meso-scale and larger scales in that their influence in general can be assessed with respect to mean winds and wind resources by corrections based on empirical relationships and simple physical models [4].
Local influences that must be considered include sheltering by obstacle(s) (more important at relatively low heights), speed-up effects of orography (e.g. hills, valleys), roughness and surface thermal conditions (or atmospheric stability, which is especially important for assessing the wind resource at higher heights above the surface). Chapter 2 will elaborate on these effects in terms of analytical models and engineering applications, while Chapter 3 is dedicated to numerical wind flow modelling. The more theoretical roots of those effects will be presented later in Chapter 9 and Chapter 10.
In most areas of the world, the local wind resource is essentially determined by large synoptic scale wind systems (they determine the overall long-term trend of a wind resource) with smaller scale features superimposed on them. The study of large scale wind systems requires long-term wind measurement, whereas local in situ wind measurements can be as short as merely one year. Therefore, long-term correlation of wind data is to some extent the study of the correlation between local effects and large scale systems. Chapter 6 will elaborate on this topic in detail.
Wind resource is a statistic quantity. Thus, the assessment of wind resource starts with the statistics of local wind speed and direction. In an overwhelming majority of cases, an on-site met mast(s) must be installed in order to reduce uncertainties of the wind resource assessment and turbine load calculations. The number of necessary masts depends on the complexity of the site. The measurement height should be as close to the expected wind turbine hub height as possible. Measuring the wind at multiple heights is necessary in order to evaluate local influences on the wind better. Due to the difficulty of constructing met masts in many cases (e.g. offshore), ground-based remote sensing (lidar and sodar) capable of measuring wind up to hundreds of metres high have been deployed more widely in recent years. Wind measuring techniques will be introduced in Chapter 8.
1.4 Uncertainty of Wind Resource
The complexity of wind resource assessment to a great extent is due to the uncertainty of the wind in nature and the uncertainty of the tools we use to predict it. It is thus something we should always keep in mind. The characteristics of statistics itself also imply that the wind, as a statistical quantity of stochastic variables, must bring in various intrinsic uncertainties as well.
The random nature of wind resources makes wind energy commercially unique to traditional fossil fuels. The price of fossil fuels fluctuates substantially with the market, which is risky for the business, whereas wind is free and never changes in price. Therefore the revenue generated by a wind power plant is not affected much by the unpredictable energy market; instead it tends to grow in the long run as electricity price generally follows an upward trend over years. On the other hand, the risk of investing in wind energy lies on how well we can estimate the wind resource and calculate the power production before erecting wind turbines and investing real money. This explains the significant role wind resource analysis and micro-siting plays for delivering a commercially successful wind power project.
To help us understand the uncertainty of wind resource better, let us take a look at a well-known example of wind speed variations at a point in Ireland, as shown in Figure 1.4. The line of monthly mean wind speed in the upper frame in Figure 1.4 indicates violent fluctuations. Yearly mean wind speeds calm down significantly, but still vary dramatically from year to year, shown more clearly by the slash blocks in the lower frame in Figure 1.4. Even averaged over 10 years, the means still do not seem to be stabilised. Noting the fact that the energy content of wind is proportional to the third power of the mean wind speed, one can imagine the magnitude of the variation in available wind energy due to fluctuating wind conditions. For example, a 1% decrease in the mean wind speed can be expected to yield about 2% less energy.
Figure 1.4 Reanalysis data for a point in Ireland showing the variability of the mean wind speed
Source: Risø National Laboratory, created by Gregor Giebel
