Renewable and Sustainable Energy Reviews 21 (2013) 131–141

Contents lists available at SciVerse ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Assessment of offshore wind energy at Younggwang in Korea

Myung Eun Lee a, Gunwoo Kim b, Shin-Taek Jeong c,n, Dong Hui Ko c, Keum Seok Kang d
a
Department of Civil and Environmental Engineering, Seoul National University, Seoul 151-742, Republic of Korea
b
Department of Ocean Civil & Plant Construction Engineering, Mokpo National Maritime University, Mokpo, Jeollanamdo 530-729, Republic of Korea
c
Department of Civil and Environmental Engineering, Wonkwang University, Iksan, Jeollabukdo 570-749, Republic of Korea
d
KEPCO Research Institute, Daejeon 305-380, Republic of Korea

a r t i c l e i n f o abstract

Article history: This study aimed to assess the potential of offshore wind power generation at Younggwang in Korea
Received 15 October 2012 which is candidate site of the offshore wind farm planned to be constructed till 2019. Toward this
Received in revised form purpose, offshore meteorological tower called HEMOSU-1 was installed and has collected the wind and
29 December 2012
other meteorological data since October, 2010. The Weibull model was fitted to the frequency
Accepted 29 December 2012
Available online 26 January 2013
distribution of data and it was found that the measured data fit well to Weibull models with properly
estimated parameters. Winter was found to be the best for harnessing wind energy because of the high
Keywords: wind speeds and consistent wind direction from the northwestern Siberian high. An assessment of the
Offshore wind energy available power density at the height of 97.35 m indicates that there is an abundant wind power
HEMOSU-1
density—averaged to be 429.20 W/m2 annually and 509.97 W/m2 during winter. After testing various
Weibull distribution
commercial wind turbines with rated power in the range of 3–7 MW, REpower 5 M was chosen for the
Wind power potential
Wind turbine analysis of installation at the wind farm in Younggwang. It was found that the annual energy
production is expected to be about 7096 GWh/yr.
& 2013 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
1.1. Situation of Korean wind power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
1.2. Offshore wind energy farm in Younggwang, Korea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2. Analysis of wind data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
2.1. Meteorological mast description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
2.2. Wind speed distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
2.3. Wind direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
2.4. Vertical wind profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
2.5. Weather condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
3. Potential power resource. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
3.1. Wind power density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
3.2. Wind turbine production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

1. Introduction depletion of fossil fuels, accompanied by unstable oil prices. In

Korea where the production of electric energy was mainly
Renewable energy is attracting rapidly increasing attention dependent on fossil fuels for a long time, there is also an
worldwide. This interest has been precipitated by increasing increasing interest in renewable energy. The necessity of having
environmental problems such as global warming and by the a stable energy supply was first realized after the two oil crises of
the 1970s. A systematic plan for developing national renewable
energy began in the 1980s, accompanied by a social recognition of
n
Corresponding author. Tel.: þ82 63 850 6714; fax: þ82 63 850 6792. the importance of having a stable energy supply. Initially, the
E-mail address: [email protected] (S.-T. Jeong). hydropower obtained from dams was the primary resource for

1364-0321/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.rser.2012.12.059
132 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141

renewable energy production. However, because suitable sites for the corresponding possible maximum electric supply was
the dam construction were limited and dam construction gave 77,180 MW; thus, the electric power reserve rate was only 5.5%
rise to negative environmental side effects, the construction of at that time. The severely cold temperature leads to a significant
hydropower dams has not progressed. Although wind power is a increase in the use of electric heating that exceeds the amount of
typical renewable and sustainable energy source, free from air-conditioning used in summer, because of relatively large cost
pollutants such as carbon dioxide and radioactive substances, it of heating fuels. Therefore, it is necessary to prepare measures to
has not been considered as a major source for national energy deal with future increase in demand in winter associated with the
supply in Korea. However, due to increasing demands for renew- climate change.
able energy and recent developments in technology, the wind The first record of a wind turbine in Korea is that of a
power becomes one of the most promising renewable energy 2.2 kW—turbine installed in Hwasung City in 1975. In 1992, a
resources. 250 kW wind turbine was installed on Jeju island; it was the first
turbine to be connected to the power grid system in Korea. The
1.1. Situation of Korean wind power first large-scale wind energy farm was established in Yeongdeok
District in 2005; it had an installation capacity of 39.6 MW in
Due to the strong effect of East Asian monsoon, Korea has four 2005. Thereafter, the construction of the wind energy farms
distinct seasons: spring, summer, autumn, and winter. The rapidly increased, due to the economical growth of Korea which
contrast between winter and summer is considerable—winter is induced the interest in sustainable development of the country.
bitterly cold, influenced primarily by the Siberian air mass from By 2009, the total installed capacity was increased up to 342 MW.
the northwest, whereas summer is hot and humid due to the The accumulated installed capacity of wind power plants from
maritime Pacific high of the southeast region. Therefore, there is 2004 to 2009 is presented in Fig. 1. It can be seen that the capacity
large fluctuation in the national electric power demand; the peak drastically increased in 2006 and 2009. Most of wind energy
demand occurs in summer or winter. Recently, due to the farms in Korea constructed during this period have only been
frequent occurrence of far below—average temperature in winter built onshore locations, in spite of the high energy potential of
associated with the abnormal climate change worldwide, handling offshore wind power plants because of the lack of detailed
the steep growth of electricity demand in winter in Korea became a information on offshore wind energy plants and other technical
matter of priority. Especially between 2010 and 2011, the maximum problems. However, there has been a continuous effort to find the
power demand increased in winter due to the abnormally intense proper site for a large-scale offshore wind farm by Korean
cold wave, whereas the annual maximum demand usually occurred government. As a result, the construction of an offshore wind
in summer before 2008. The most recent maximum electric energy farm is planned in Younggwang as introduced in the next
demand was recorded to be 73,140 MW on January 17, 2011 and section.

400
342
350
Capacity (MW)

300
232 37
250
:Wind farm
200 168 172
:Substation
150 :Meteorological tower
:Electric grid
100 Saemankum S/S
58 345 kV
Latitude (N)

50 18
0 80 km, HVDC
2004 2005 2006 2007 2008 2009 36

3rd Stage: 1st Stage:

Fig. 1. Accumulate total installed capacity in South Korea. Large Scale Verification

Gochang
2nd Stage: 22 km, HVAC PT center S/S:
Demonstration 154 kV
HEMOSU-1 Younggwang

35

125 126 127 128

Longitude (E)

Fig. 3. Locations of offshore wind farm and electric grids to substations with
bathymetry contour for water depth (m).

Table 1
The impact of depth and distance on costs [3].

Water Distance from shore (km)

depth (m)
0–10 10–20 20–30 30–40 40–50 50–100 100–200 4200

10–20 1 1.02 1.04 1.07 1.09 1.18 1.41 1.60

20–30 1.07 1.09 1.11 1.14 1.16 1.26 1.50 1.71
30–40 1.24 1.26 1.29 1.32 1.34 1.46 1.74 1.98
40–50 1.40 1.43 1.46 1.49 1.52 1.65 1.97 2.23
Fig. 2. Map of wind energy farm locations.
 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141 133

1.2. Offshore wind energy farm in Younggwang, Korea Saemankum substation is 80 km, which means its cost premium
factor is about 1.26 in Table 1. There is another economic factor to
The Korean government is planning to construct a 2500 MW- be considered for the construction of wind farm in the offshore of
class offshore wind farm by 2019 along the south-western coast. Younggwang because of the flimsy ground of sea bed around this
The selected site is in the Younggwang region (Fig. 2), where the site. On the flimsy sea bed, multi-pile structure such as jacket
wind is of Class 3 (6.9–7.5 m/s), the mean sea water depth is type or pile cap is preferred to mono-pile with gravity type as the
20 m, and the distance of the site from the substation is about support structure of wind tower, and the cost of multi-pile
15 km. The outline of the strategic plan for the establishment of structure is generally larger than mono-pile. Providing 3 MW-
this offshore wind energy farm is divided into three phases: turbine on the 15 m water depth, a jacket type structure results in
more cost than pile cap through the economic evaluation [4].
 First stage: by 2014, construct a 100 MW test bed for verification. Therefore, pile cap method for multi-pile support is recom-
 Second stage: by 2016, construct a 400 MW demonstration site. mended for the construction of wind farm.
 Third stage: by 2019, establish a 2000 MW large-scale wind farm. In addition to economic feasibility of wind farm, there are
several environmental factors to be considered for the construc-
The private and public sectors will invest a total of 10.2 trillion tion of wind farm around Younggwang. The south-western part of
won (92.7 billion USD) to build a large-scale offshore wind farm Korea, in which Younggwang is located, is on the East Asian–
with 2500 MW. The location of the wind energy farm, i.e., Australasian Flyway which is the one of the important pathways
Younggwang, was selected after examining the Korea National of a migrant. Various kinds of birds use the coastal region of
Wind Map, which was established by using the RDAPS (Regional south-western part of Korea as their route or settlement. Thus, it
Data Assimilation and Prediction System) data with a grid size of is necessary to evaluate the effect of the offshore wind farm on
3 km and a frequency of 3 h from 2005 to 2007 [1]. RDAPS is an the moving path of birds through this region. Furthermore, the
operational weather forecast model based on the MM5 model [2]. south-western coastal region is also well-known for its shallow
The roadmap of the offshore wind farm around Younggwang is water depth and tidal flat with great tidal range. Due to this
presented in Fig. 3. In the first and second stages, test bed will be geographical characteristic, a fishery and aquaculture industry is
connected to Gochang substation with grid of 154 kV—HVAC by the main source of income in this area. Therefore, the research for
using domestic technology. In the third stage, the large-scale the preservation of fishing ground here and the agreement with
wind farm will be connected to Saemankum substation of 345 kV fishermen with respect to fishery activity is prerequisite before
using HVDC transmission. the construction of wind farm. These environmental problems
The costs for offshore wind generation are usually much mentioned above will be testified after construction of the test
higher than for onshore wind farms because of the additive cost bed for verification in the first stage.
for transmission grid to substations inland and their maintenance Towards the final stage of wind farm construction in Younggwang
cost against corrosion and meteorological damage. The level of region, deficient infrastructure for electric grid connection in the
the cost premium of offshore wind relative to onshore wind south-western part of Korea also has to be resolved. After demonstra-
stations depends on factors including water depth and distance tion of the second stage, large-scale offshore wind farm is planned to
from shore. EEA [3] provided the adjustment factor by which be connected to Saemankum substation using HVDC. However, it is
investment and installation costs should be multiplied for deeper not possible to connect only with domestic technical skill, and
water and greater distance as shown in Table 1. According to construction by a foreign company results in relatively high cost
Fig. 3, the location of wind farm of the third stage in plan around and low contribution to national technical level. Furthermore,
Younggwang is inside 30 m isobaths and the distance to government will continue to expand the facility of Younggwang

Fig. 4. HEMOSU-1 offshore meteorological unit.

134 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141

Table 2
Description of facilities in HEMOSU-1.

El. (m) (above MSL) Equipment El. (m) (above MSL) Equipment

97.35 Anemometer 56.31 Anemoscope, anemometer

96.31 Anemoscope, anemometer 46.31 Anemoscope, anemometer
92.85 Barometers, thermo-hygrometers 26.31 Anemometer
86.31 Anemometer 13.05 Barometers, thermo-hygrometers
76.31 Anemoscope, anemometer 11.72 Rainfall sensor
66.31 Anemometer 10.02 Deque

Table 3 10
Annual mean, standard deviation, and maximum wind speed of measurement in 9
2011 by HEMOSU-1.
8

Wind speed (m/s)

Height (m) Mean Standard Max. 10 min 7
(m/s) deviation (m/s) average (m/s)
6
97.350 7.122 3.907 30.570 5
96.310 6.942 3.900 30.760 4
86.310 6.853 3.750 30.120
76.310 6.752 3.703 29.760 3
66.310 6.818 3.718 29.970 2 height 97.35 m
56.310 6.505 3.608 28.850
46.310 6.394 3.451 27.950 1 height 26.31 m
26.310 6.081 3.301 26.730 0
Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.

Fig. 5. Annual variation of wind speed in Younggwang.

Table 4
Seasonal mean, standard deviation, and maximum wind speed of measurement in
2011—winter by HEMOSU-1. Table 5
Estimated annual Weibull parameters.
Height Mean Std. deviation Max. 10 min
(m) (m/s) (m/s) average (m/s) Height 2011 2011—winter

97.350 7.614 3.856 24.460 k c k c

96.310 7.413 3.878 24.580
86.310 7.326 3.684 24.070 97.350 1.919 8.029 2.093 8.597
76.310 7.238 3.667 23.860 96.310 1.870 7.819 2.021 8.367
66.310 7.362 3.734 24.060 86.310 1.924 7.725 2.109 8.272
56.310 6.999 3.666 22.930 76.310 1.920 7.611 2.093 8.173
46.310 7.000 3.543 22.220 66.310 1.932 7.688 2.090 8.312
26.310 6.874 3.514 21.380 56.310 1.897 7.330 2.018 7.899
46.310 1.954 7.211 2.095 7.904
26.310 1.942 6.858 2.073 7.761

offshore wind farm after completion of the third stage. The capacity of depicted in Table 2 and Fig. 4. The sampling rate of data is 0.3 s
existing grid facility in this area is far from sufficient to manage larger and they are averaged at a regular interval of 10 min.
than a 1500 MW-power plant. Extensive supplement of electrical grid
in south-western region of Korea should be carried out for the 2.2. Wind speed distribution
successful settlement of power production of the Younggwang
large-scale offshore wind farm. The statistical information of annual data is presented in
This study provides the assessment of the possible energy Table 3. The annual mean of wind speed was observed as
production by the wind farm at Younggwang based on the wind 6.081–7.122 m/s which increases as the observation elevation
data of 1 yr in 2011 measured near the candidate site. becomes high. To investigate the winter data, observation
data during January, February, and December were separately
collected, and the statistical characteristics during these months
are presented in Table 4. The mean wind speed in winter was
2. Analysis of wind data about 0.5 m larger than the annual mean wind speed.
To investigate the annual variation of wind speed, the monthly
2.1. Meteorological mast description mean wind speeds of year 2011 are plotted in Fig. 5. The maximum
mean wind speed of 8.609 and 7.539 m/s was observed in January at
The wind measurement at a specific location can effectively the top and bottom observation elevations of 97.35 m and 26.31 m,
represent the wind characteristics of the site under consideration. whereas the minimum mean wind speed of 6.202 and 5.158 m/s was
For this reason, KEPCO (Korea Electric Power Corporation) observed in October and June, respectively. The annual fluctuation
installed the HEMOSU-1 (Herald of Meteorological and Oceano- range of wind speed was about 2.4 m/s between maximum and
graphic Special Unit-1), an offshore tower (Fig. 4) at 1261 070 minimum monthly averaged wind speed.
45.3000 E and 351 270 55.1700 N, in order to measure the wind data To evaluate the wind energy potential of a site, it is convenient
in the Younggwang region. The height of the tower is about 100 m to introduce the probability distribution of the site’s wind speed.
above mean sea level, and its equipments and their elevations are Weibull distribution has been generally used in many literatures
 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141 135

0.14 0.14
Measurement data Measurement data
0.12 0.12
Weibull distribution Weibull distribution

Probability density
Probability density 0.1 0.1

0.08 0.08

0.06 0.06

0.04 0.04

0.02 0.02

0 0
0 5 10 15 20 25 0 5 10 15 20 25
Wind speed (m/s) Wind speed (m/s)

0.14 0.14
Measurement data Measurement data
0.12 0.12
Weibull distribution Weibull distribution
Probability density

Probability density
0.1 0.1

0.08 0.08

0.06 0.06

0.04 0.04

0.02 0.02

0 0
0 5 10 15 20 25 0 5 10 15 20 25
Wind speed (m/s) Wind speed (m/s)

Fig. 6. Histograms of wind speed data and estimated Weibull distributions in 2011: (a) 26.31 m, (b) 46.31 m, (c) 76.31 m, (d) 96.31 m.

0.14 0.14
Measurement data Measurement data
0.12 0.12
Weibull distribution Weibull distribution
Probability density

Probability density

0.1 0.1

0.08 0.08

0.06 0.06

0.04 0.04

0.02 0.02

0 0
0 5 10 15 20 25 0 5 10 15 20 25
Wind speed (m/s) Wind speed (m/s)

0.14 0.14
Measurement data Measurement data
0.12 0.12
Weibull distribution Weibull distribution
Probability density

Probability density

0.1 0.1

0.08 0.08

0.06 0.06

0.04 0.04

0.02 0.02

0 0
0 5 10 15 20 25 0 5 10 15 20 25
Wind speed (m/s) Wind speed (m/s)

Fig. 7. Histograms of wind speed data and estimated Weibull distributions in 2011—winter: (a) 26.31 m, (b) 46.31 m, (c) 76.31 m, (d) 96.31 m.
136 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141

to express the wind speed frequency distribution of a specific where V and s are the mean and the standard deviation of the
location [5–10]. In terms of Weibull distribution, the probability observation data V, respectively, such as
density function for the wind speed V is expressed as X
N
  "   # V¼ Vi ð5Þ
k V k1 V k i¼1
f ðV Þ ¼ U Uexp  ð1Þ
c c c
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u N
uX  2
where k is the dimensionless shape parameter, and c is the scale s¼t V i V =ðN1Þ ð6Þ
parameter. The cumulative distribution function, which gives the i¼1
probability that the wind velocity, is equal or lower than V is
"   # where Vi is the wind speed averaged for every 10 min, and N is the
V k total number of data for a given duration.
F ðV Þ ¼ 1exp  ð2Þ
c The parameters k and c were obtained from the annual and
winter data, respectively. Estimated parameters of each observa-
Analytical or empirical methods can be used for the parameter tion height for annual and winter are shown in Table 5. The
estimation, and the method of Justus et al. [11,12] was used in maximum k is obtained as 2.109 at 86.31 m in winter, whereas
this study as the maximum c was obtained to be 8.597 at 97.35 m in winter.
 1:086 The both adjusted Weibull models of annual and winter data
s approach to the Rayleigh distribution for k¼2. However, k of
k¼ ð3Þ
V annual distribution is smaller than 2 whereas that of winter
2:6674
slightly exceeds over 2. The larger values of k in winter imply that
Vk the wind speed distribution in winter is less positively skewed
c¼ 2:73855
ð4Þ
0:184 þ 0:816k than those of annual data.

N
NNW NNE
N
NNW NNE
NW NE

NW NE

WNW ENE
Unit : m/s
Unit : m/s >0 - 5
WNW ENE
>0 - 5 >5 - 7

>5 - 7 >7 - 10
W E
0% 5% 10% 15% 20% 25% 30% >10 - 12
>7 - 10
W E
0% 5% 10% 15% 20% >10 - 12 >12 - 15

>12 - 15 >15
WSW ESE
>15
WSW ESE

SW SE

SW SE
SSW SSE
S
SSW SSE
S
N
N NNW NNE
NNW NNE
NW NE
NW NE

WNW ENE
Unit : m/s
WNW ENE
Unit : m/s >0 - 5
>0 - 5 >5 - 7
>5 - 7 >7 - 10
W E
>7 - 10 0% 5% 10% 15% 20% 25% >10 - 12
W E
0% 5% 10% 15% 20% >10 - 12 >12 - 15
>12 - 15
>15
WSW ESE
>15
WSW ESE

SW SE
SW SE

SSW SSE
SSW SSE S
S
Fig. 9. Wind rose diagram of measurement in 2011—winter: (a) 46.31 m,
Fig. 8. Wind rose diagram of measurement in 2011: (a) 46.31 m, (b) 96.31 m. (b) 96.31 m.
 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141 137

The parameter c is definitely small in the low elevations, 30

because it is deeply related to the mean wind speed. As shown
13 m
in Table 4, the decreasing rate of c in winter is smaller than that of 25
annual distribution, which implies that the variation of wind 93 m

Temperature (oC)
speed is relatively small in winter season. 20
Figs. 6 and 7 show the wind data at elevations of 26.31, 46.31,
76.31, 96.31 m, and their corresponding Weibull distributions for 15
the annual and winter season, respectively. Weibull seems to give
a reasonable fit to observed distributions of both annual and 10
winter data. Due to the relatively small number of the observation
in winter, the histograms of measurement showed large fluctua- 5
tions in the range of the middle wind speed in Fig. 7. However, the
mean distribution of observation data is not largely deviated from 0
Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.
the estimated Weibull, and it fits well especially in the high
speed range. Fig. 11. Monthly mean values of temperature measured by HEMOSU-1 in 2011.

2.3. Wind direction 1030

1025
It is well known that the meteorological characteristics of the
1020
Korean peninsula are dominated by the seasonal monsoon. The

Pressure (mb)
north-western wind induced by Siberian high is dominant through- 1015
out Korea in the winter season, while the south-southeastern wind 1010
from the Pacific high dominates in summer. In spring and autumn, 1005
the north-northeastern winds comprise a relatively large propor- 1000
tion of the wind, and the south-southwestern wind is added in 995
autumn. Because Younggwang is located on the western coast of 990
the Korean peninsula, this site is vulnerable to the influence of the 13 m
985
winter monsoon with prevailing northwestern wind. Furthermore, 93 m
980
the winds in winter are much stronger than those in any other
Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.
season.
In the measurement of this study, wind directions were Fig. 12. Monthly mean values of atmospheric pressure measured by HEMOSU-1
measured at four elevations: 46.31, 56.31, 76.31, 96.31 m as in in 2011.
Table 2. Among them, data measured at 46.31 and 96.31 m
1.30
1
1.25
0.9
Density (kg/m3)

1.20
0.8
log V

0.7 log V= 0.111log z + 0.622 1.15

1.10
0.6 26.31 m
97.35 m
0.5 1.05
Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.
1 1.2 1.4 1.6 1.8 2 2.2
log z Fig. 13. Computed monthly mean values of air density of Younggwang.

1
were depicted as wind roses in Figs. 8 and 9 for the annual
and the winter, respectively. As shown in Fig. 8, north-
0.9 northwestern (NNW) and south-southeastern (SSE) winds domi-
nated, respectively, in winter and summer. Fig. 8 clearly shows
0.8 the strong monsoon effect on the seasonal wind direction.
log V

log V = 0.0699logz+ 0.733 As shown in the winter wind rose depicted in Fig. 9, the wind
0.7 of NW, NNW, and N is greatly prevalent, which comprises about
54%
0.6 and 49% of the wind during the winter at 46.31 and 96.31 m,
relatively.
0.5
1 1.2 1.4 1.6 1.8 2 2.2 2.4. Vertical wind profile

log z
It is important to ensure the accuracy of the wind data at the
Fig. 10. Vertical profile of the measured wind speed: (a) for annually averaged level of the turbine blade. In most meteorological station, wind
data, (b) for averaged data during winter. data are collected at a certain elevation, but we measured the
138 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141

wind speed at eight different elevations, which provide the 429.20

typical shape of vertical wind profile around the candidate site. 97.35
509.97
According to the literature [5,7,13,14], the most commonly 407.03

Measurement Height (m)

96.31
used method to adjust the wind velocity at one level to another is 484.62
the power law method expressed as [15,16] 86.31 384.42
452.27
 b
z 76.31 371.07
V ¼ V0 ð7Þ 440.51
z0
66.31 380.22
where V0 is the wind speed recorded at the anemometer height, 464.99
z0, and b is the roughness factor of the site. In this study, b was 56.31 337.73
obtained by applying least square method to the linear relation 413.15
between logz and logV. Fig. 10 shows the calculated VðzÞ at eight 46.31 315.82
400.85 annual
observation elevations in log–log axis. Vertical profile of annually
averaged data was linearly approximated by b ¼0.111 while the 26.31 276.82 winter
384.61
profile of averaged wind during winter has gradient of b ¼ 0.0699.
0 100 200 300 400 500 600 700
Wind power density (W/m2)
2.5. Weather condition
Fig. 16. Annual mean power density and seasonal mean power density in winter
from measurement of HEMOSU-1.
Because of seasonal variation of prevailing air-mass, meteor-
ological characteristics such as temperature and atmospheric
pressure also show large variations for each season. Fig. 11 shows
the monthly mean values of temperature measured at elevations Fig. 12 shows the monthly mean atmospheric pressures
of 13.05 and 92.85 m except for January when meteorological measured at 13 m and 93 m. Due to the relatively high variations
data were not collected. Temperature showed large variations of the seasonal temperature, the difference between the max-
ranging from 25.5 1C in July and 3.4 1C in February at the low imum and minimum monthly-averaged pressure at upper eleva-
elevation (13 m). tion (93 m) is observed about 27.3 mb. The vertical stratification
of atmospheric pressure in summer was negligibly small, whereas
the difference in pressure in February reached up to 12.7 mb.
Dec. Relatively low temperature and high pressure in winter make
427.38
the wind energy potential much higher than that of other season
Nov. 276.95
because the air density is proportional to atmospheric pressure
Oct. 213.43
and inversely proportional to temperature by the following ideal
Sep. 212.69
gas law
Aug. 258.07
Jul. 216.82 p
r¼ ð8Þ
Jun. 205.47 RT
May 202.73
where R is ideal gas constant with 287 J/kg K, and pand T are,
Apr. 263.81
respectively, the monthly mean pressure and temperature.
Mar. 336.86
Monthly mean air density at Younggwang was calculated with
Feb. 229.58
measurement data and it is presented in Fig. 13. Air density also
Jan. 472.46
shows larger vertical variation in winter compared to summer.
0 100 200 300 400 500 600
Monthly power density (W/m2)- 26.31 m
3. Potential power resource
Fig. 14. Monthly power density from measurement at 26.31 m of HEMOSU-1.

3.1. Wind power density

Dec. 525.00 The potential wind power density P available over a unit area
Nov. 375.52 normal to a wind of speed V is represented by
Oct. 265.16 1
Sep. 289.58 P¼ rV 3 ð9Þ
2
Aug. 356.00
Jul. 480.60 The mean wind power density can be calculated using the
Jun. 405.93 measured data at each observation height by the following
May 500.51 equation [13]:
Apr. 520.74 N obs
1 X
Mar. 499.31 P¼ r n V3 ð10Þ
2N i ¼ 1 i i
Feb. 334.33
Jan. 673.04 where r is the average air density for the given duration, Vi is the
0 100 200 300 400 500 600 700 800 wind speed at the mid-point of the ith class, ni is the frequency of
Monthly power density (W/m2)- 97.35 m occurrence in the ith class. Monthly mean power densities are
calculated based on the time series of measurement of each
Fig. 15. Monthly power density from measurement at 97.35 m of HEMOSU-1. month at height 26.31 and 97.35 m and presented in Figs. 14
 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141 139

and 15, respectively. At two elevations, the peak monthly power 3.2. Wind turbine production
density was obtained in January.
In Fig. 16, the annually averaged power density and seasonally For a given probability density function of the wind speed and
averaged values for winter were compared at each elevation. The turbine power curve, Pt(V), the average wind turbine power Pt is
annually averaged power density was obtained as 429 W/m2 and obtained from
the seasonal power density in winter was obtained as 510 W/m2 Z 1
at the top measurement elevation. Pt ¼ P t ðVÞf ðVÞ dV ð12Þ
0
There are several classifications that indicate the qualitative With a summation over NB, the number of bins, the following
evaluation of wind power density. Mirhosseini et al. [9] and expression, can be used to find the average wind machine power P [9]
Manwell et al. [17] introduced the following categories in terms
XNB
of annual average wind power density: 1         
Pt ¼ V i þ 1 V i f V i þ 1 Pt V i þ 1 þ f V i P t V i ð13Þ
i¼1
2
 P o 100 W=m —low. 2

 P  400 W=m2 —good. In the following, some kinds of commercial offshore wind
 P 4 700 W=m2 —great. turbines are selected which have nominal power higher than
3.0 MW and available hub height near 97.35 m, to evaluate the
net energy production at Younggwang region. Six wind turbines
Another classification by the EWEA (European Wind Energy selected for the comparison are presented in Table 7 along with
Association) for the wind power density is as follows [18]: their rated power and other characteristics. All six selected are
horizontal-axis, variable speed wind turbines with pitch control.
 P  300400 W=m2 —fairly good.
 P  500600 W=m2 —good.
 P  700800 W=m2 —very good. 8000
REpower 5M
7000 REpower 6M
According to the above criteria, the wind power in Fig. 16 6000 Siemens SWT-3.6-107
presents a good condition for the construction of wind farm if the
Power output (kW)
Vestas V90-3.0MW
turbine hub height is above 46.31 m. 5000
Vestas V112-3.0MW
We also calculated P using the estimated seasonal Weibull 4000 Vestas V164-7.0MW
models in Table 5 with the following relationship [7]:
Z 1 3000
1 1  
P¼ rV 3 f ðV Þ dV ¼ rc3 G 1þ 3=k ð11Þ
0 2 2 2000

For annual and winter distributions, the power density calcu- 1000

lated from Eqs. (10) and (11) is compared at each measurement 0

elevation, and their absolute errors are presented in Table 6 which 0 5 10 15 20 25
shows that the estimated Weibull distribution accurately esti- Wind speed (m/s)
mates the power density from real distribution of measurement.
Fig. 17. Power curves of selected machines.

Table 6
Annual energy production (MWh/yr)

30,000 0.5
Estimated wind power density from measurement and fitted Weibull model E_A C_f
(W/m2).
25,000 0.401
0.4
0.341
Height 2011 2011—winter

Capacity factor
0.324
20,000 0.314 0.306
0.292 20,909 0.3
Measure. Weibull Err. Measure. Weibull Err. 15,740
15,000 14,198

10,534 0.2
26.31 276.82 272.06 4.76 384.61 385.79 1.18 10,000
9,917
8,029
46.31 315.82 313.36 2.46 400.85 401.88 1.03
56.31 337.73 340.00 2.27 413.15 415.19 2.04 0.1
5,000
66.31 380.22 383.33 3.10 464.99 466.35 1.37
76.31 371.07 374.18 3.12 440.51 442.14 1.63 0 0.0
86.31 384.42 389.57 5.15 452.27 453.98 1.71 REpower 5M REpower 6M SWT-3.6-107 V90-3.0MW V112-3.0MW V164-7.0MW
96.31 407.03 417.02 9.99 484.62 489.37 4.75
97.35 429.20 437.86 8.67 509.97 512.58 2.61 Fig. 18. Annual energy production and capacity factor of selected machines
in 2011.

Table 7
Characteristics of selected wind turbines.

Model Rated Diameter (m) Hub Cut-in wind Rated wind Cut-out wind Class (IEC)
output (kW) height (m) speed (m/s) speed (m/s) speed (m/s)

REpower 5 M 5000 126 90–100 3.5 13.5 30 IB

REpower 6 M 6150 126 85–95 3.5 15 30 IB
Siemens SWT-3.6-107 3600 107 80–96 3–5 13–14 25 IA
Vestas V90-3.0 MW 3000 90 84–119 3.5 15 25 IIA
Vestas V112-3.0 MW 3000 112 84–119 3 12 25 IIA
Vestas V164-7.0 MW 7000 164 Site specific 4 13.5 25 S
140 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141

The capacity factor CF is also an important index in measuring

Energy production during winter

8,000 0.5
E_A C_f 0.457
7,000 the productivity of a wind turbine. It compares the actual power
0.395
0.376
6,000
0.364
0.355
0.4 production of a turbine over a given time duration with the
0.341

Capacity factor
(MWh/3 months)

5,972 amount of power of the turbine operated at the rated output for
5,000 0.3
4,064
4,531 the same time duration [10,22]
4,000
2,959 0.2 EA
3,000 2,829 CF ¼ ð15Þ
2,302 ER
2,000
0.1
1,000
where ER(¼ PRT) is the maximum power of the turbine operating
at the rated output PR for time duration T.
0 0.0
REpower 5M REpower 6M SWT-3.6-107 V90-3.0MW V112-3.0MW V164-7.0MW The net energy production of each turbine and its capacity
factor during 2011 and winter of 2011 is presented in Figs. 18
Fig. 19. Seasonal energy production and capacity factor of selected machines in and 19. At least 8000 MWh/yr of energy production is possible by
2011—winter.
using one of the listed wind turbines in Table 7. During winter, we
can expect at least 4000 MWh of energy production using wind
turbine with output higher than 5 MW.
Because the master plan of the wind farm in the Younggwang
6 determined to install 5 MW class wind turbines to be tied in a plant,
the energy production by Repower 5 M turbine was selected to be
5 2011
analyzed in detail. The power duration curve is useful for comparing
2011- winter the energy potential by choosing the wind turbine type. It is a graph
Power (MW)

4
with the power output of a specific turbine on the y-axis and the
3 number of hours in the given time duration for which the speed
equals or exceeds each particular value on the x-axis [17]. Fig. 20
2 shows the power duration curve of REpower 5 M obtained from the
measurement wind data at an elevation of 97.35 m. Because the areas
1 under the curves are proportional to the available wind energy, the
proportion of the energy potential in winter compared to annual
0
0 1000 2000 3000 4000 5000 6000 7000
production is visually apparent in this power duration curve. The time
duration for this turbine to maintain the output as higher than 3 MW
Duration (h)
is about 550 h in winter, which is about 31% of annual time duration
Fig. 20. Duration curves of REpower 5 M. for output higher than 3 MW.
As we introduced the plan for wind farm around Younggwang
region, it is planned to be installed about 500 machines with
The pitch control system is one of the most widely used control 5 MW capacity for the future construction of 2500 MW-level
techniques to regulate the output power of a wind turbine wind farm. For REpower 5 M, about 810 MW (P t  500) of
generator. The method relies on the variation in the power annually averaged energy supply is expected for the complete
captured by the turbine as the pitch angle of the blades is wind farm with 2011 wind data. During the winter of 2011,
changed. At wind speeds below the rated power of the generator, 940 MW of energy supply by 500 machines of Repower 5 M was
the pitch angle is at its maximum though it can be lower to help estimated. As a result, the completed wind farm is expected to
the turbine accelerate faster. Above the rated wind speed, the provide about 12% of the peak electric demand in winter, and
pitch angle is controlled to keep the generator power at rated increases the electric power reserve rate by 1.3% in terms of 2011
power by reducing the angle of blades. electricity supply condition; the maximum electric demand of
Among the selected turbine, Siemens’ SWT 3.6-107 and Vestas 73,140 MW in January 2011. The total annual energy production
V90 are established models in the offshore wind turbine market, in this wind farm is expected to be 7096 GWh/yr.
whose total installed capacity is 270 and 290 MW, respectively
[19]. A 3 MW-level turbine has been the leading large capacity
models until recently. However, due to the recent developments 4. Conclusion
in large capacity wind turbines, higher rated turbines such as the
REpower 5 M/6 M are expanding their market share rapidly. In this research, the offshore wind energy potential of the wind
Furthermore, the latest technical progress in offshore turbines farm, which is planned to be set up in Younggwang located on the
has led to the commercialization of machines with enlarged west coast of the Korean peninsula, was evaluated. For the
capacity, such as Vestas’ new V-164-7.0 MW launched in March, assessment, measurement data of 2011 from a meteorological
2011. Design parameters for wind class provided by IEC [20] of mast called HEMOSU-1 are obtained; installed at 1271 070 45.3000
each selected turbine are also presented in Table 7. Because 50 yr E and 351 270 55.1700 N, a location of offshore near Younggwang:
extreme wind speed around Younggwang was estimated 39.8 m/s
[21], wind turbine class upper than class II is possible to use in (1) The annual mean wind speed for this site is estimated to be
this site. 7.122 m/s at an elevation of 97.35 m, which is relatively much
The wind observation data at 97.35 m was selected for the stronger than the mean wind speed for any onshore location
assessment of the net energy production using Eq. (12). Power in Korea.
curves of each turbine are presented in Fig. 17. The net energy (2) The Weibull probability distribution was estimated from the
production EA for a given duration T, 365 days for an year and 90 frequency distribution of measurement data at each observation
days for winter season, is obtained by using the following elevation and statistically significant results were obtained.
equation [22]; (3) From the measurement data, it was observed that the north-
northwestern wind direction is prevalent due to the winter
EA ¼ P t T ð14Þ
monsoon.
 M.E. Lee et al. / Renewable and Sustainable Energy Reviews 21 (2013) 131–141 141

(4) Vertical profile of annually averaged data at Younggwang was [3] European Environment Agency. Europe’s onshore and offshore wind energy
linearly approximated by the power law with power of b ¼0.111. potential: an assessment of environmental and economic constraints, EEA
technical report no. 6/2009. European Environment Agency; 2009.
(5) The assessment of the available power density at 97.35 m [4] Kang KS, Lee JS, Kim JY, Ryu MS. Economic analysis of offshore wind farm
indicates that there is a high wind power density averaged to considering domestic development conditions of Korea. Journal of Wind
be 429.20 W/m2 annually and 509.97 W/m2 during winter. Energy 2011;2(1):37–43 in Korean.
[5] Ahmed Shata AS, Hanitsch R. The potential of electricity generation on the
This agrees well with the value obtained from the Weibull east coast of Red Sea in Egypt. Renewable Energy 2006;31:1597–615.
parameters’ investigation. [6] Ahmed Shata AS. Wind energy as a potential generation source at Ras Banas,
(6) Among the selected commercial wind turbines with rated Egypt. Renewable and Sustainable Energy Reviews 2010;14:2167–73.
[7] Dahmouni AW, Ben Salah M, Askri F, Kerkeni C, Ben Nasrallah S. Assessment
power in the range of 3–7 MW, REpower 5 M was analyzed in of wind energy potential and optimal electricity generation in Borj-Cedria,
terms of the power duration curve. The results showed that Tunisia. Renewable and Sustainable Energy Reviews 2011;15:815–20.
the expected annual energy production will be about [8] Fyrippis I, Axaopoulos PJ, Panayiotou G. Wind energy potential assessment in
Naxos Island, Greece. Applied Energy 2010;87:577–86.
7096 GWh/yr when the construction of the wind energy farm
[9] Mirhosseini M, Sharifi F, Sedaghat A. Assessing the wind energy potential
is completed as planned. locations in province of Semnan in Iran. Renewable and Sustainable Energy
(7) Based on the maximum electric demand of 73,140 MW in Reviews 2011;15:449–59.
January 2011, the completed wind farm is expected to cover [10] Mostafaeipour A, Sedaghat A, Dehghan-Niri AA, Kakantar V. Wind energy
feasibility study for city of Shahrbabak in Iran. Renewable and Sustainable
12% of peak demand and increase the electric power reserve Energy Reviews 2011;15:2545–56.
rate by about 1.3% in the winter season. [11] Justus CG, Mikhail A. Height variation wind speed and wind distributions
statistics. Geophysical Research Letters 1976;3(5):261–4.
[12] Justus CG, Hargraves W, Garvine R. Methods for estimating wind speed
frequency distributions. Journal of Applied Meteorology 1978;17:350–3.
Acknowledgments [13] Ilinca A, McCarthy E, Chaumel JL, Retiveau JL. Wind potential assessment of
Quebec Province. Renewable Energy 2003;28:1881–97.
This work is the outcome of a Manpower Development [14] Omer AM. On the wind energy resources of Sudan. Renewable and Sustain-
able Energy Reviews 2008;12:2117–39.
Program for Marine Energy by the Ministry of Land, Transport [15] Awanou CN, Degbey JM, Ahlonsou E. Estimation of mean wind energy
and Maritime Affairs(MLTM) and a study on the demonstration available in Benin (Ex Dahomey). Renewable Energy 1991;1(5/6):845–53.
project for 2.5 GW offshore wind farm at the southern part of [16] Khogali A, Albar OF, Yousif B. Wind and solar energy potential in Makkah
(Saudi Arabia)_comparison with Red Sea coastal sites. Renewable Energy
yellow sea by the Ministry of Knowledge Economy(MKE).
1991;1(3):435–40.
[17] Manwell JF, McGowan JG, Rogers AL. Wind energy explained: theory, design
and application. Amherst, USA: John Wiley & Sons; 2002.
References [18] Garrad A. Wind energy in Europe: a plan of action, summary report of wind
energy in Europe-time for action. The European Wind Energy Association; 1991.
[19] Andreas T Concepts for high power wind turbines introducing HTS technol-
[1] Kim JY, Kang KS, Oh KY, Lee JS, Ryu MS. Assessment of possible resources and ogy. In: World green energy forum, Gyeongju, Korea; 2010. p. 17–9.
selection of preparatory sites for offshore wind farm around Korean Peninsula. [20] IEC. Wind turbines—part 1: design requirements. IEC 61400-1; 2005.
Journal of the Korean society for new and renewable energy 2009;5(2):39–48 in [21] Oh KY, Kim JY, Lee JK, Ryu MS, Lee JS. Wind resources assessment and
Korean. performance evaluation of wind turbines for the test-bed of offshore wind
[2] Warner TT, Kuo YH, Doyler JD, Dudhia J, Stauffer R, Seaman NL. Nonhydro- farm. Journal of Wind Energy 2011;2(1):15–20 in Korean.
static, meso-beta-scale, real-data simulations with the Penn State University/ [22] Chang T-J, Wu Y-T, Hsu H-Y, Chu C-R, Liao C-M. Assessment of wind
National Center for Atmospheric Research mesoscale model. Meteorological characteristics and wind turbine characteristics in Taiwan. Renewable Energy
Atmospheric Physics 1992;49:209–27. 2003;28:851–71.