An Assessment of Offshore Wind Energy Potential Using Mesoscale Model and GIS
Atsushi YAMAGUCHI* Takeshi ISHIHARA*
and Yozo FUJINO**
*Institute of Engineering Innovation, School of Engineering, The University of Tokyo 2-11-16 Yayoi Bunkyo TOKYO 113-8656 JAPAN **Department of Civil Engineering, School of Engineering, The University of Tokyo 5-8-1 Hongo Bunkyo TOKYO 113-8656 JAPAN
e-mail.
[email protected]
Abstract: Offshore wind energy potential in the supply area of Tokyo Electric Power Company (TEPCO) was investigated by using mesoscale model and geographical information system (GIS). Following results were obtained. 1) Wind climate predicted by mesoscale model shows good agreement with the observation and the prediction error of annual mean wind speed was 4.8%. 2) Annual mean wind speed offshore Choshi is 7.5m/s while at northern site, it decreases to 5.7m/s. This is because Choshi is located at the tip of a cape. 3) Concerning the economical and social criteria, the available potential becomes 94TWh/year, accounting for 32%of the annual demand of TEPCO. 4) If bottom mounted foundation is used, 0.4TWh/year of energy will be exploited, which accounts for only 0.1% of the annual demand of TEPCO. Key Words: Offshore wind energy potential, Mesoscale model, Geographical information system
1. Introduction A large portion of wind energy potential in Japan is located at rural area, where demand is low and the grids are weak and the limitation in integrating wind energy to the grid exists. On the other hand, in the supply area of Tokyo Electric Power Company, which is the biggest utility in Japan and supplies electricity 282TWh per year, onshore wind resource is limited and little land is left for large scale wind farm although there is no limitation on grid connection of wind energy. Thus, if offshore energy potential around the supply area of TEPCO is estimated, certainly it will help the penetration of wind energy in Japan. A few studies have been carried out so far on the estimation of offshore wind energy potential around Japan. Nagai et al.1) used wind speed date observed at lighthouses around Japan and Fuji2) used satellite SSM/I data to estimate offshore wind energy potential around Japan. In both studies the distance to shore is limited to 3km although water depth is not considered. Henderson et al.3) also used satellite data to estimate the offshore resource around Japan considering water depth. Distance to shore and the water depth is not the only factor to limit the area for the construction of wind farm. For example fishery rights are often said to limit the offshore wind energy development in Japan. However, it is not clear how much of the offshore energy is available excluding the area with
fishery rights are established. For onshore wind resource assessment, it is a common practice to account for such social constraints. Voivontas et al.4) used geographic information system (GIS) and developed a decision support system for wind resource assessment named RES-DSS considering social criteria. Hillring and Kreig5) also used GIS to estimate the wind energy potential excluding densely populated area and military area. These studies showed that GIS is a powerful tool to consider the social criteria and estimate the wind energy potential. In this study, the offshore wind resource potential around the supply area of TEPCO is estimated considering economical and social criteria by using mesoscale model and GIS.
2. Wind Climate Assessment by Mesoscale Model To investigate the wind climate and its spatial distribution in the offshore of TEPCO supply area, simulation by Mesoscale Model was carried out for the year 2000. 2.1 Model As a mesoscale atmospheric model, RAMS (Regional Atmospheric Modeling System)6) was used in this study. RAMS is based on non-hydrostatic Reynolds-averaged primitive equations. The governing equations are as follows:
• Equation of motion ∂u ∂u ∂u ∂u ∂π ' = −u −v −w −θ + fv ∂t ∂x ∂y ∂z ∂x
∂u ⎞ ∂u ⎞ ∂ ⎛ ∂ ⎛ ∂u ⎞ ∂ ⎛ ⎟+ ⎜ Km ⎜ Km ⎟ ⎜ Km ⎟+ ∂z ⎠ ∂y ⎟⎠ ∂z ⎝ ∂x ⎝ ∂x ⎠ ∂y ⎜⎝ ∂v ∂v ∂v ∂v ∂π ' − fu = −u −v − w −θ ∂t ∂x ∂y ∂z ∂y +
∂ ⎛ ∂v ⎞ ∂ ⎛ ∂v ⎞ ∂ ⎛ ∂v ⎞ ⎜⎜ K m ⎟⎟ + ⎜ K m ⎟ ⎜ Km ⎟ + ∂x ⎝ ∂x ⎠ ∂y ⎝ ∂y ⎠ ∂z ⎝ ∂z ⎠ ∂w ∂w ∂w ∂w ∂π ' gθ v ' = −u −v −w −θ − ∂t ∂x ∂y ∂z ∂z θ0 +
∂ ⎛ ∂w ⎞ ∂ ⎛ ∂w ⎞ ∂ ⎛ ∂w ⎞ ⎟⎟ + ⎜ K m ⎟ + ⎜⎜ K m ⎜ Km ⎟ ∂x ⎝ ∂x ⎠ ∂y ⎝ ∂y ⎠ ∂z ⎝ ∂z ⎠ • Thermodynamic equation ∂θ il ∂θ ∂θ ∂θ ∂θ ⎞ ∂ ⎛ = −u il − v il − w il + ⎜ K h il ⎟ ∂x ⎠ ∂t ∂x ∂y ∂z ∂x ⎝ +
radiation, soil and vegetation is also included in the model. RAMS adopts a rotated polar-stereographic projection as the horizontal grid and the z terrain–following coordinate system as the vertical grid.
2.2 Computational Domain Two level nested grids were used to simulate the offshore wind climate. Figure 1 shows the location of each grid. A grid with horizontal resolution of 2km (Grid 2) was set to cover the offshore area around TEPCO supply area. To increase the computational efficiency, two separate grid were used as grid 2. One is located at the offing of Choshi (Grid 2 Choshi), and the other is located at Sagami Bay (Grid 2 Sagami). To take the effect of surrounding mountain areas into account, grid 1 with horizontal resolution of 8km was set around grid 2 as shown in figure 1.
∂θ ⎞ ∂ ⎛ ∂θ ⎞ ⎛ ∂θ ⎞ ∂ ⎛ ⎜⎜ K h il ⎟⎟ + ⎜ K h il ⎟ + ⎜ il ⎟ ∂y ⎝ ∂y ⎠ ∂z ⎝ ∂z ⎠ ⎝ ∂t ⎠ rad • Water species mixing ratio equation +
∂rn ∂r ∂r ∂r = −u n − v n − w n ∂t ∂x ∂y ∂z +
•
supply area of TEPCO
∂ ⎛ ∂rn ⎞ ∂ ⎛ ∂rn ⎞ ∂ ⎛ ∂rn ⎞ ⎟ + ⎜ Kh ⎜ Kh ⎟ + ⎜ Kh ⎟ ∂y ⎟⎠ ∂z ⎝ ∂x ⎝ ∂x ⎠ ∂y ⎜⎝ ∂z ⎠
Grid 1
Mass continuity equation
Rπ 0 ∂π ' =− ∂t c v ρ 0θ 0
Grid 2 (Choshi)
⎛ ∂ρ 0θ 0 u ∂ρ 0θ 0 v ∂ρ 0θ 0 w ⎞ ⎜⎜ ⎟ + + ∂y ∂z ⎟⎠ ⎝ ∂x
Grid 2 (Sagami)
Table I. symbols used in this paper
θil rn
ρ rad g rt rv
π π’
definition east-west wind component north-south wind component vertical wind component Coriolis parameter eddy viscosity coefficient for momentum eddy viscosity coefficient for heat and moisture ice-liquid water potential temperature water mixing ratio species of total water, rain, pristine crystals, aggregates, and snow air density subscript denoting tendency from radiation parameterization gravity total water mixing ratio water vapor mixing ratio total Exner function perturbation Exner function
Symbols used in the above equations are shown in table I. Further, turbulent kinetic energy parameterization by Mellor and Yamada level 2.5 scheme and the parameterization of convection,
Figure 1. The supply area of TEPCO and the computational domain of mesoscale model 2.3 Verification of the model For the verification of estimated wind climate, predicted wind speed and wind direction are compared with the observation at Choshi meteorological station. 15 wind speed (m/s)
symbol u v w f Km Kh
Observation
Model
10
5
0
0
30
60
90
120 150 180 210 240 270 300 330 360 Day
F igure 3. Day averaged wind speed at Choshi meteorological station in 2000 Figure 3 shows day averaged wind speed at Choshi meteorological station. The predicted value shows good agreement with the observation and prediction
error of the annual mean wind speed was 4.8%.
annual
June N
W
E 10%
N
W
E 20%
15%
S
Model
October N
W
E 20%
30%
S
The reason for this is that when the wind blows from SSW, which is one of the prevailing wind direction, the are north to Choshi is located behind the land. This makes the wind speed there relatively lower while Choshi is located at the tip of the cape and no decrease in wind occurs for any prevailing wind direction.
30%
S
Observation
Figure 2. Wind rose at Choshi meteorological station in 2000: a) annual; b) June; c) October The predicted wind direction also shows good agreement. The windrose at Choshi meteorological station for annual, June and October is shown in figure 4. The seasonal variation of prevailing wind direction that in June the prevailing wind direction is south southwesterly while it is north northeasterly to northeasterly in October, is well predicted by mesoscale model.
n a e m l a u n n a
) s/ 8 m ( d7 e e p s d6 ni w 5
Choshi -40
-20
South
0
20 y (km)
40
60
80
North
Figure 4. Annual mean wind speed along the line located at 10km from the coastline
3. Geographic Information System 3.1 Social criteria
2.3 Spatial distribution of annual mean wind speed Annual mean wind speed differs considerably depending on the location even offshore. Figure 3 shows the spatial distribution of annual mean wind. Generally wind speed decreases as the distance from the shore decreases. However, at some areas wind speed is considerably lower than the other area. Figure 4 shows the annual mean wind speed along the line located at 10km from the coastline, which is indicated in white line in figure 3. At the offing of Choshi, corresponds to 0km on y-axis, the annual mean wind speed reaches 7.5m/s although at the northern area annual mean wind speed is much lower. For example, at the area 80km north to Choshi, annual mean wind speed is only 5.7m/s and where the energy density is only 44% of that at the offing of Choshi.
In some offshore areas, it will be prohibited to exploit the wind energy due to social criteria. Since the exploitation of wind energy in these areas highly depends on political and social decision, it is assumed that no energy will be exploited in these areas in this study. Table 1 summarizes the social criteria concerned in this study. All of the Geographical data mentioned in table 1 are provide by Ministry of Land, Infrastructure and Transportation Japan (MLIT). Areas with fishery rights, national parks, areas within 10km from the coastline and port areas were excluded from the exploitation of wind energy.
Table 1. Social criteria concerned in this study Areas with fishery rights National Parks Distance from the coastline
Ports
Choshi
Line located at 10km from the coastline
Figure 3. Distribution of Annual mean wind speed and the line located at 10km from the coastline
Excluded from the exploitation of energy
Excluded from exploitation of energy
the
Areas within 10km from the coastline are excluded from
the exploitation of energy due to the landscape reason. Excluded from the exploitation of energy
These social criteria concerned in this study are shown in figure 5. Ports and national parks occupy relatively small areas and their location is limited. Thus, they will not have significant effect on total available potential. On the other hand, areas with fishery rights and areas excluded by landscape reason will have substantial effect. Fishery rights are established at almost all the coastlines making it impossible to exploit wind energy near shore even when the landscape criteria is to be eased. At some areas, fishery rights are established even if the distance from the coastline is more than 10km.
Water depth (m) Ibaraki
Chiba
40km from the coastline
Sagami Bay
Figure 5. Social criteria considered in this study
Figure 6. Economical criteria considered in this study
3.2 Economical criteria It will be difficult to exploit wind energy at some areas due to economical and technical criteria even within the areas where no social criteria are applied. In this study, criteria shown in table 2 were considered as economical criteria. Water depth data is provided by Marine Information Research Center of Japan Hydrographic Association.
4. Available wind energy potential
Water depth
Distance from the coastline
Areas where water depth is more 20m is excluded for bottom mounted foundations and more than 100m is excluded for floating foundations. Areas where the distance to the coastlines is longer than 40km is excluded from the exploitation of
wind energy. Water depth is an important factor. For bottom-mounted foundation, which is widely used for offshore development so far in Europe, bathymetry must be shallower than 20m. If floating foundation, which is still under development all over the world, can be used, the limitation of the bathymetry will be 100m and more areas can be used for wind farm as discussed by Henderson et al.3). Still, area where bathymetry is deeper than 100m is difficult to develop. Distance from shore is related to transmission and maintenance cost. The value of 40km was used in this study. The areas with these economical criteria are shown in figure 6. At the east offshore of Chiba and Ibaraki water depth is relatively shallow. On the other hand, at Sagami Bay, seabed is very steep and even within 40km from the coastline water depth exceeds 1500m at some sites.
4.1 Wind turbine To estimate the available potential, wind turbine model has to e assumed. In this study 2MW wind turbines are assumed to be settled with the interval of 8D by 8D. The detail of the turbine and the assumed arrangement is summarized in table 3 and figure 7. 2000 output (kW)
Table 2. Economical criteria concerned in this study
Total available potential in this offshore area, spatial distribution and the available potential for each water depth class was calculated concerning the social and economical criteria.
1500 1000 500 0
0
5
10 15 wind speed (m/s)
20
25
Figure 7. Power curve of wind turbine assumed in this study Table 3. Wind turbine assumed in this study Rated power Rotor diameter Hub height Spacing
2MW 80m 60m 8D×8D
4.2 Total available potential and its spatial distribution To help the decision making process of the location and design of offshore wind farm, total available potential and its spatial distribution in this area were
calculated. The floating foundation was considered and maximum water depth was set to 100m. Figure 8 shows the available potential per square kilometers in this area. Apparently most of the available potential is located at the east offshore of Chiba and Ibaraki especially at the offing of Choshi while available area around Sagami Bay is limited. This is mainly due to the seabed topography as discussed in Chapter 3.2.
Ibaraki
5. Conclusion In this study, offshore wind energy potential in the supply area of Tokyo Electric Power Company (TEPCO) was investigated considering social and economical criteria by using mesoscale model and geographical information system (GIS). Following results were obtained. 1)
Predicted annual mean wind speed, day averaged wind speed and wind direction show good agreement with observation. The prediction error of annual mean wind speed was 4.8%.
2)
Annual mean wind speed offshore Choshi is 7.5m/s while at northern site, it decreases to 5.7m/s although the distance from the coastline is same. This is because Choshi is located at the tip of a cape,
3)
Concerning all the economical and social criteria, the available potential becomes 94TWh/year, accounting for 32%of the annual demand of TEPCO.
4)
If bottom mounted foundation is used, 0.4TWh/year of energy will be exploited, which accounts for only 0.1% of the annual demand of TEPCO.
Choshi
Chiba Sagami Bay
References
Figure 8. Distribution of available potential By integrating this, total available potential in this area becomes 94TWh per year, which accounts for 32% of the annual demand for TEPCO. 4.3 Available potential for each water depth class To clarify how much of this total available potential can be exploited by bottom mounted foundation, available potential for each water depth class was calculated and shown in figure 8. If bottom mounted foundation, which is common in Europe, is used, only 0.4TWh per year will be exploited, which accounts for only 0.1% of the annual demand of TEPCO. On the other hand, 68% of the total available potential is located in the area where water depth is between 20 and 200m. This implies that floating foundation should be developed to exploit offshore wind energy in this area.
1)
Nagai, H., Ushiyama, I., Ueno and Y.: Feasibility study of predicting offshore wind resources and power generation in Japan., Proc. Wind Energy Symposium, Vol. 19, pp. 168-171, 1997. (in Japanese)
2)
Fujii, T.: An estimation of the potential of offshore wind power in Japan by satellite data. Proc. Japan Solar Energy Society / Japan Wind Energy Association Joint Conference, paper 155, pp. 25-26, 1999.
3)
Henderson, A. R., Leutz, R. and Fujii, T.: Potential for floating offshore wind energy in Japanese waters, Proc. Twelfth International Offshore and Polar Engineering Conference, pp. 505-512, 2002.
4)
D. Voivontas, D. Assimacopoulos and A. Mourelatos: Evaluation of Renewable Energy Potential Using a GIS Decision Support System, Renewable Energy, Vol. 13, No. 3, pp. 333-344, 1998.
5)
Hillring, B. and Krieg, R.: Wind Energy Potential in Southern Sweden – Example of Planning Methodology, Renewable Energy, Vol. 13, No. 4, pp. 471-479, 1998.
6)
Pielke, R. A. et al.: A Comprehensive Meteorological Modeling System – RAMS, Meteorol. Atmos. Phys., Vol. 49, pp. 69-91, 1992.
available potential (kWh/year)
50 39
40 30
25
20 12 10 0
9
9
300-400
400-500
0.4 0-20
20-100
100-200 200-300 water depth (m)
Figure 8. Available potential for each water depth class
