OFFSHORE WIND RESOURCE ASSESSMENT, SITE SUITABILITY AND TECHNOLOGY SELECTION FOR BLIGH WATERS FIJI USING WINDPRO
Dissertation in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE WITH A MAJOR IN ENERGY TECHNOLOGY WITH FOCUS ON WIND POWER
Uppsala University
Department of Earth Sciences, Campus Gotland
KUNAL KUSHAL DAYAL
MAY 2015
OFFSHORE WIND RESOURCE ASSESSMENT, SITE SUITABILITY AND TECHNOLOGY SELECTION FOR BLIGH WATERS FIJI USING WINDPRO
Dissertation in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE WITH A MAJOR IN ENERGY TECHNOLOGY WITH FOCUS ON WIND POWER
Uppsala University
Department of Earth Sciences, Campus Gotland
Approved by:
Supervisor, Dr. Sasan Sarmast
Examiner, Professor Jens Nørkær Sørensen
Date, June 2015
ABSTRACT
This thesis aims to carry out offshore wind resource assessment, site suitability and technology selection for Bligh Waters in Fiji, and perform energy calculations for a 10-12 MW model offshore wind farm as well as carry out a simple economic analysis.
The objectives were achieved by assessing the offshore wind resources employing atmospheric reanalysis data from the WindPRO online database performing the data correlation using the Measure Correlate Predict (MCP) module of WindPRO. The best correlated wind speed closest to the microsite was computed to be about 6.5 m/s at a height of 10 m, with a dominant East-southeast (ESE) and South-southeast (SSE) wind directions. Furthermore, the wind turbine technology was selected to be Vestas V80- 2.0MW Offshore wind turbine and Siemens SWT-2.3-93 (2,300 kW) wind turbine with wind turbine class IEC IA for the site using the analysis done by the WindPRO site compliance module.
Moreover, energy calculations were performed for 10 MW and 11.5 MW model offshore wind farms using the best correlated datasets close to the micro-site. The best model offshore wind farm was found to be the 11.5 MW wind farm, which had an annual energy production of 40,327.5 MWh/year, a capacity factor of 40.0 %, park efficiency of 99.8 % and full load hours of 3507 hours/year. Each Siemens SWT-2.3-93 (2,300 kW) wind turbine of the 11.5 MW wind farm produces an average of 8,065.4 MWh annually.
The wind farm has a simple payback time of approximately 8 years with an installation
cost of USD $51,750,000 and AAR of USD $6,452,400. The cost of energy generation
per kWh is computed to be USD $0.12. Thus, comparing this to the cost of energy
generation by other renewable and conventional sources in Fiji, it can be concluded that it
is feasible and potentially competitive to invest into offshore wind farms to support the
national electricity grid in Fiji.
ACKNOWLEDGEMENTS
This thesis would not have been possible without the generous support of numerous people. Firstly, I would like to thank my Sponsor Erasmus Mundus Kite Partnership for providing me the scholarship to pursue this Master of Science degree at Uppsala University Campus Gotland in Visby, Sweden. Secondly, I would like to thank my supervisor Dr. Sasan Sarmast for his guidance, time and support he provided for my thesis.
Thanks also go to my friends and colleagues and the department faculty and staff for making my time at Uppsala University Campus Gotland a great experience.
Finally, thanks to my mother and father for their encouragement support and love.
NOMENCLATURE
AAR Average Annual Return
AEP Annual Energy Production
COAMPS Coupled Ocean/Atmospheric Mesoscale Prediction System
FDOE Fiji Department of Energy
FEA Fiji Electricity Authority
FJD Fijian Dollar (currency)
MCP Measure Correlate Predict
MIUU Meteorological Institute of Uppsala University
MW Megawatt
MWh Megawatt hours
GW Gigawatt
GWh Gigawatt hours
IEC International Electro-technical Commission
km/hr Kilometres per hour
kW Kilowatt
kWh Kilowatt hours
m Metres
MEASNET Measuring Network of Wind Energy Institutes
m/s Metres per second
rpm Revolutions per minute
USD United States Dollar (currency)
V Volts
W/m
2Watts per square meter
WD Wind Direction
WRF Weather Research and Forecasting
WS Wind Speed
WTG Wind Turbine Generator
TABLE OF CONTENTS
Page
ABSTRACT ... iii
ACKNOWLEDGEMENTS ... vi
NOMENCLATURE ... v
TABLE OF CONTENTS ... vi
LIST OF FIGURES ... vii
LIST OF TABLES ... ix
1 INTRODUCTION AND BACKGROUND ... 1
1.1 PREVIOUS WORK IN THIS FIELD ... 4
1.2 OBJECTIVES OF THE THESIS ... 5
2 THEORY ... 6
2.1 WIND AND ITS CHARACTERISTICS ... 6
2.2 WIND TURBINES ... 9
2.3 PREDICTION OF WIND SPEED AND DIRECTION ... 11
2.4 SITE SUITABILITY ASSESSMENT ... 12
2.5 WIND TURBINE DESIGN CLASS ... 13
2.6 ECONOMIC ANALYSIS ... 14
3 SIMULATIONS AND SETUP ... 17
3.1 SIMULATION PROCEDURE ... 17
3.2 ECONOMIC ANALYSIS ... 27
4 RESULTS AND DISCUSSION ... 29
5 CONCLUSION AND FUTURE WORK ... 43
REFERENCES ... 45
APPENDIX ... 48
LIST OF FIGURES
Page
Figure 1 Global Cumulative Installed Wind Capacity from 1997 – 2014 ... 1
Figure 2 Map of Fiji showing Bligh Waters and the Micro-Site for the Model Offshore Wind Farm ... 2
Figure 3 Fiji Electricity Authority Electricity Generation for 2013 ... 3
Figure 4 Schematic View of a Horizontal and a Vertical Axis Wind Turbines ... 10
Figure 5 Schematic View of the Horizontal Axis Wind Turbine for Offshore Applications and Foundation Options ... 11
Figure 6 Location of Reanalysis Datasets from WindPRO Online Database ... 20
Figure 7 Local Short-Term Data and Long-Term Reference Data for the Best Datasets close to the Micro-Site ... 21
Figure 8 Linear Regression Analysis of Wind Speed for Site (short-term) with Long Term Reference ... 22
Figure 9 Regression MCP Wind Speed Prediction of Site Wind Data with Long Term Reference ... 23
Figure 10 Matrix MCP Analysis of Wind Speed and Wind Direction at Reference Position ... 24
Figure 11 Matrix MCP Wind Speed Prediction of Site Wind Data with Long Term Reference ... 24
Figure 12 Radar Diagram of the Mean Wind Speed and Frequency at Bligh Waters In Fiji ... 30
Figure 13 Plot of Wind Roses for Most of the Wind Resources Assessed Showing Dominant Wind Direction ... 31
Figure 14 Diurnal Mean Wind Speed of Bligh Waters in Fiji ... 33
Figure 15 Diurnal Mean Wind Direction of Bligh Waters in Fiji ... 33
Figure 16 Annual Mean Wind Speed of Bligh Waters in Fiji ... 34
Figure 17 Annual Mean Wind Direction of Bligh Waters in Fiji ... 34
Figure 18 Site Compliance Results of the 10 MW Model Wind Farm Using Vestas
Wind Turbines in Bligh Waters Fiji ... 36 Figure 19 Site Compliance Results of the 11.5 MW Model Wind Farm Using
Siemens Wind Turbines in Bligh Waters Fiji ... 37 Figure 20 Calculated Annual Energy Production of the 10 Model Wind Farm in
Bligh Waters Fiji ... 38 Figure 21 Calculated Annual Energy Production of the 11.5 Model Wind Farm in
Bligh Waters Fiji ... 39
LIST OF TABLES
Page Table 1 Wind Shear Coefficient of Various Terrains ... 8 Table 2 Quality of Reference for Correlation Coefficients ... 12 Table 3 Specifications for Wind Turbine Design Class ... 14 Table 4 Present Costs of Electricity Generation in Fiji by Different Technology . 16 Table 5 Best Correlation Results of Wind Speed and Wind Direction using
Linear Regression and Matrix MCP Methods ... 29 Table 6 Wind Shear Coefficients using Mean Wind Speed from EmdERA
E178.562 S17.193 Dataset at Different Heights ... 32 Table 7 Wind Turbine Technology Selected for Bligh Waters, Fiji ... 35 Table 8 Economic Analysis of the 10 MW Model Offshore Wind Farm using
Vestas Wind Turbines in Bligh Waters, Fiji ... 40 Table 9 Economic Analysis of the 11.5 MW Model Offshore Wind Farm using
Siemens Wind Turbines in Bligh Waters, Fiji ... 41
Table 10 Summary of Wind Resources Datasets (1-17) in Bligh Waters Fiji ... 48
Table 11 Summary of Wind Resources Datasets (18-34) in Bligh Waters Fiji ... 49
CHAPTER 1 INTRODUCTION AND BACKGROUND
Wind energy is a renewable energy resource that is available almost everywhere on earth which has been used for various purposes in the ancient times from sails to propel ships and boats and later as grain grinding mills and water pumps for mankind. In addition, the energy in the wind can also be used to rotate wind turbines to produce electricity. The first modern electricity producing wind turbine was made in 1890 in Denmark to electrify rural areas (Mathew, 2006). Wind energy is a fast growing industry for power generation with a worldwide installed capacity of 318 GW as of the year ending 2013 (REN21, 2014) and 369.553 GW as of the year ending 2014 (GWEC, 2015) and many more assessments, permits and installations are underway. In Figure 1, it can be seen that there has been a tremendous increase in the global cumulative installed wind capacity by almost a factor of 10 from 2003 to 2014.
Figure 1 Global Cumulative Installed Wind Capacity from 1997 – 2014. (Source: GWEC, 2015)
This thesis aims to model an offshore wind farm in Fiji. Fiji is located in the western South Pacific Ocean between the latitudes of 12 °S - 22 °S and longitudes of 177
°E – 178 °W. There are more than 332 islands with a total land area of approximately
18,400 km
2. Only 110 of the islands are inhabited. The two largest islands are Viti Levu
and Vanua Levu which take up 87 % of the total land area. The two islands are mountainous and of volcanic origin with maximum peaks of 1300 m. Fiji has a tropical climate with a dry and wet season. The wet season extends from November to April while the dry season is from May to October.
The location of the micro-site for this study is in Bligh Waters which is located in between the two larger islands in Fiji. The wind resources in terms of online atmospheric reanalysis datasets in and around this region have been assessed. The map in Figure 2 clearly shows the coverage of the region of study by the highlighted grid area and the location of the micro-site for the model offshore wind farm which is located 1 kilometer from the shore and in water depth of less than 30 meters.
Figure 2 Map of Fiji showing Bligh Waters and the Micro-Site for the Model Offshore Wind Farm
(Source: Google Earth).
Fiji like other renewable energy focused countries have also installed a wind farm onshore in Butoni, Sigatoka which consists of 37 Vergnet wind turbines (Model: GEV- MP 275) each rated 275 kW making a total installed capacity of 10 MW to support its national electricity grid. Figure 3 presents the electricity generation of Fiji for the year 2013. Looking at the electricity generation statistics of Fiji for the year ending 2013 it can be stated that renewable energy based power plants produced 63 % of electricity while fossil fuel based power plants produced 37 % of electricity. The 10 MW wind farm produced 1 % towards the total electricity generation from renewables while 60 % comes from hydro-power plants and 2 % from independent power producers (IPPs) (FEA Annual Report, 2013). Fiji has numerous onshore locations where there is potential for wind power development to support the national electricity grid as well as to support rural electrification in places where there is no grid access like interiors and outer islands (Fiji Department of Energy, 2015). Research has been carried out by researchers in Fiji by using the data provided by NASA’s Solar System Exploration (SSE) and Atmospheric Data Center (ADC) and this, has shown that the average yearly wind speeds for Fiji is between 5 to 6 m/s with an average power density of 160 W/m
2(Kumar and Prasad, 2010).
Figure 3 Fiji Electricity Authority Electricity Generation for 2013 (Source: FEA Annual Report, 2013).
Hydro 527.397 60%
Wind 5.348 1%
IPPs 14.719
2%
Thermal 324.755 37%
Fiji Electricity Authority Electricity Generation 2013 [GWh]
Hydro Wind IPP Thermal
1.1 PREVIOUS WORK IN THIS FIELD
A number of researches have been done in Fiji using measured wind data from the FDOE onshore wind monitoring program (Fiji Department of Energy, 2015). In a prefeasibility study of wind resources by Singh (2015) in Vadravadra on Gau Island in Fiji it has been reported that the annual wind speed over southern Gau varies from 8.42 to 14.69 m/s and the power density at a height of 50 m or higher is found to be an annual average of around 1128 W/m
2. WAsP analysis of total energy produced using eight Vergnet 275 kW wind turbine generators was 13.320 GWh and the COE borne was FJD
$0.55/kWh (USD $0.29).
In a study of wind energy potential, resource assessment and economics in Qamu, Navua, Fiji it has been reported that average wind speed at 30 m was 4.60 m/s. The wind speed at 55 m height was calculated to be 6.31 m/s with a power density of 300 W/m
2and an annual energy production of 677 MWh using one Vergnet 275 kW wind turbine using WAsP analysis. The cost of electricity generation was calculated to be FJD 0.08/kWh (USD $0.04) with a payback period of approximately 10 years (Kumar and Nair, 2014).
It has been reported that average wind speed at 30 m is 6.24 m/s with a mean power density of 590 W/m
2in another study of wind power potential at Benau, Savusavu, Fiji.
Using WAsP analysis and two wind turbines Vestas V27 and Vergnet 275 kW at the site, the mean annual electricity production is calculated to be 641 MWh per turbine. The levelised cost of energy was calculated to be FJD $0.08/kWh (USD $0.04) and an internal rate of return of 21.3 % (Kumar and Nair, 2013).
In a feasibility study of offshore wind energy potential in Kijal, Malaysia using QuikSCAT satellite data from WindPRO database wind resources have been assessed and the economic efficiency have been evaluated by means of the expected capacity factor.
Seven different sizes of wind turbines ranging from 110 kW to 1250 kW have been used
and it was reported that the 850 kW wind turbine was the best wind turbine for installations
at the site in terms of its best capacity factor of 26.8 % (Ibrahim et al., 2014). So far, no research has been performed specifically for offshore wind resources to support the national electricity grid in Fiji.
1.2 OBJECTIVES OF THE THESIS
The major reason for this study is to outline the feasibility of offshore wind resources to support the electricity sector as in Fiji hydro-power dominates in power generation while there is only one onshore wind farm of 10 MW. A number of studies have been done onshore for wind resources but mainly aimed at rural electrification in rural and outer islands where there is no grid access and there has been no research done specifically to explore offshore wind resources to support the national electricity grid. The results of this study will create offshore wind resources knowledge about Bligh Waters in Fiji and also assist investment decision making in wind power development offshore by interested government, local and international private sector investors.
This study aims to assess the offshore wind resources, plot wind roses to determine the dominant wind direction and carry out data correlation of the wind speed and wind direction data from various sources of atmospheric reanalysis data available online WindPRO database for Bligh Waters in Fiji. Select the best correlated results to represent the wind speed at the site and determine the shear coefficients. Plot diurnal and annual patterns of wind speed and wind direction for the site together with the assessment of technology based on the site characteristics. Model a 10-12 MW offshore wind farm using the appropriate wind turbine technology selected and hence carry out energy calculations and finally a simple economic analysis.
The thesis is organized as follows: In chapter 1 – 2, background, motivation and theory of
the work are presented. The procedure including the simulation setups are reported in
chapter 3. In chapter 4, the results are presented and discussed. Conclusions and future
work of the thesis are presented in chapter 5.
CHAPTER 2 THEORY
2.1 WIND AND ITS CHARACTERISTICS
The sun is the original source of energy that generates the earth’s renewable wind resource. There is uneven heating of the earth by solar radiation which causes temperature differences and thus atmospheric pressure differences across the earth’s surface which generates wind (Manwell et. al, 2009). Wind is the movement of air from a high pressure to a low pressure region. The driving forces on the air parcels in the atmosphere are gravitational force, pressure gradient force, Coriolis force, Centrifugal force and friction force. The balance of these forces in the vertical and horizontal directions gives rise to the different kinds of winds one of which is taken into account by the wind simulation software’s as geostrophic wind.
An air parcel which is at rest initially will move from a high pressure region to low pressure region due to the pressure gradient force. When the air parcel moves it is deflected by the Coriolis force (due to the rotation of the earth) to the right in the northern hemisphere and vice-versa in the southern hemisphere. As the air parcel gains speed, this deflection increases until the Coriolis force equals the pressure gradient force. This condition is called the geostrophic balance and at this point the air parcels are parallel to the isobars. Therefore, this movement of air parcels parallel to the isobars due to this balance is referred to as geostrophic winds. Geostrophic winds are largely driven by temperature and thus pressure differences, and are unaffected by the surface of the earth.
The geostrophic wind is found at altitudes higher than 1000 meters (Nilsson and Ivanell, 2010).
In addition, the winds that are very much influenced by the ground surface of the earth at lower altitudes up to 100 meters are called surface winds. The surface roughness and the orography of the earth’s surface will slow the wind down. When dealing with wind energy the major concerns are surface winds and the usable energy content of the wind.
The direction of the wind near the earth’s surface will be slightly different from the
direction of the geostrophic wind because of the Coriolis force which is due to the rotation of the earth (SOPAC, 2009).
The measurements of wind speed and wind direction are done using various instruments. The oldest technology and the standard one as outline by MEASNET to measure wind speed and wind direction are by using the mechanical anemometers and wind vanes at different heights above the earth’s surface at the site of interest. Latest technology makes use of SODARs (Sound Detection and Ranging), LIDARs (Light Detection and Ranging) and satellite measurements of wind speed and wind direction.
Since the measurements done by these technologies are not at the hub height of the wind turbines which is of interest, so the wind shear which is the variation of the wind speed with height is needed. It is also vital to have a better understanding of the wind shear coefficients of a particular site for wind power development as it directly impacts the available power at the hub height of the wind turbine and also affects the wind turbine blades in terms of the cyclic loadings on the blades (Ray et. al., 2006).
Furthermore, according to Manwell et al. (2009) two laws have been generally used in wind energy studies to model the vertical profile of wind speed over regions of homogeneous, flat terrain like fields, deserts and prairies. The first law is the log law namely logarithmic wind profile equation and the other one is the power law. The two equations are given below.
Logarithmic law: 𝑈(𝑧) =
𝑈∗𝑘
𝑙𝑛 (
𝑧𝑧0
) (1)
Power law:
𝑈(𝑧)𝑈(𝑧𝑟)
= (
𝑧𝑧𝑟
)
𝛼(2)
Where 𝑈(𝑧) is the wind speed at height 𝑧, 𝑈 ∗ is the frictional velocity, k is the von
Karman’s constant which equals to 0.4, 𝑧
0is the surface roughness length, 𝑈(𝑧
𝑟) is the
reference wind speed at height 𝑧
𝑟and 𝛼 is the power low exponent or the wind shear
coefficient.
Moreover, making 𝛼 which is the wind shear coefficient the subject from the power law the equation of the wind shear coefficient can be computed.
𝛼 = ln(𝑢(𝑧)) − ln(𝑢(𝑧
𝑟))
ln(𝑧) − ln(𝑧
𝑟) (3)
Many researchers have used the one-seventh power law where 𝛼 =
17
for relatively flat terrains. Likewise, Table 1 summarizes the values for the wind shear coefficients for different types of terrains as outlined by Patel (1999).
Table 1 Wind Shear Coefficient of Various Terrains.
Terrain Type Power law exponent or Wind Shear Coefficient (α)
Lake, ocean, and smooth-hard ground 0.10
Foot-high grass on level ground 0.15
Tall crops, hedges, and shrubs 0.20
Wooded country with many trees 0.25
Small town with few trees and shrubs 0.30
City area with tall building 0.40
Besides, to compute the power in the wind and the energy which is generated using a wind turbine are given by:
Power: 𝑃
𝑊𝑇𝐺=
12
× 𝜌 × 𝐴 × 𝑣
3× 𝐶
𝑝(4)
Energy: 𝐸 =
12
× 𝜌 × 𝐴 × 𝑣
3× 8760 × 𝜂 (5)
Where: 𝑃
𝑊𝑇𝐺is the power from the wind turbine in 𝑘𝑊, E is the energy output in kWh, 𝜌
is the density of air, 𝐴 is the swept area by the wind turbine blades in 𝑚
2, v is average
wind speed at the hub height of the wind turbine in 𝑚 𝑠
−1, 𝐶
𝑝is the power coefficient
which is an indicator of the efficiency of the turbine, 8760 is the total number of hours in
a year and η is the efficiency of the wind turbine.
There are analytical and numerical models that are utilized for computing wind resources. In this study, WindPRO software is used and it utilizes WAsP which is an analytical model for the AEP calculation. Wind Atlas Analysis and Application Program (WAsP) is a linear model which has been developed by Risø laboratories in Denmark. The model takes into account the following parameters:
1. The geostrophic balance, where the geostrophic drag law gives the geostrophic wind ‘G’: 𝐺 =
𝑈∗𝑘
√[ln
𝑈∗𝑓𝑧0
− 𝐴(𝜇)]
2+ 𝐵
2(𝜇) (6)
Where: A, B are dimensionless functions of stability and f is the Coriolis parameter.
2. The modified logarithmic wind profile.
𝑈(𝑧) =
𝑈∗𝑘
(ln
𝑧𝑧0
− 𝜓 (
𝑧𝐿
)) (7)
Where: 𝜓 is the stability dependent function which is positive for unstable and negative for stable conditions.
3. A specific (but uniform) stability, roughness variations and height variations.
The WindPRO PARK module uses WAsP together with several wake models and advanced turbulence computation facilities for wind farm AEP calculations.
2.2 WIND TURBINES
Wind turbines are the machines that extract energy from the wind. The two
common types of wind turbines used onshore are the horizontal axis wind turbine and the
vertical axis wind turbine. The schematic view of the two kinds of wind turbines are shown
in Figure 4 with labelled components.
Figure 4 Schematic view of a Horizontal and a Vertical Axis Wind Turbines (Source: English Eco Energy).
For offshore applications the most common wind turbine is the horizontal axis
wind turbine with a monopole foundation for water depths less than 30 m. The schematic
view with different foundation options are shown in Figure 5.
Figure 5 Schematic view of the Horizontal Axis Wind Turbine for Offshore Applications and Foundation Options (Source: VJ Tech).
2.3 PREDICTION OF WIND SPEED AND DIRECTION
The Measure Correlate Predict (MCP) method is a statistical procedure which is
utilized to predict long term wind speed and wind direction at a potential wind farm site
by comparing short term on-site wind measurements to nearby long term meteorological
stations or available long term atmospheric reanalysis datasets. There has to be concurrent
data between the short term and the long term measurement and the on-site measurements
should be for a minimum period of at least one year. WindPRO has a MCP Module which
has four calculation models for long term correction of short term on site measurements.
The four calculation models are the Linear Regression MCP, Matrix MCP, Weibull Scale MCP and Wind Index MCP.
The correlation coefficient of the MCP methods indicates the quality of the reference. Table 2 summarizes the quality of reference for the different range of correlation coefficients (WindPRO 2.9 User Manual, 2013).
Table 2 Quality of Reference for Correlation Coefficients.
Correlation Coefficient Quality of Reference
0.5 to 0.6 Very poor
0.6 to 0.7 Poor
0.7 to 0.8 Moderate
0.8 to 0.9 Good
0.9 to 1.0 Very good
2.4 SITE SUITABILITY ASSESSMENT
Site suitability is the assessment of a wind turbine class at a particular location taking into account a number of assessment parameters as per the international standards IEC61400-1 ed. 3 (2010). WindPRO has a site compliance module which calculates and evaluates seven main checks for site suitability as per the mentioned international standard. For each one of the main checks, the site compliance module evaluates whether a particular wind turbine class for instance, IEC IA complies with the actual site and layout conditions. According to EMD International A/S (2015), the site compliance module helps identify critical risks in a wind farm project and calculates the seven main checks at the hub height of individual wind turbine position as required in the international IEC 61400- 1 ed. 3 (2010) standards which follow:
1. Terrain complexity – terrain steepness and variability in the vicinity of each WTG
position.
2. Extreme wind – refers to 10 minute averaged wind speed event with a recurrence period of 50 years.
3. Effective turbulence – represents the fatigue loads, a more long-term degradation of structural integrity of the wind turbine.
4. Wind distribution – the frequency of occurrence at different wind speeds for each WTG.
5. Wind shear – the vertical variation of wind speed across the rotor for each WTG position.
6. Flow inclination – the sector with highest absolute (positive or negative) flow inclination for each WTG.
7. Air density – the density of air at hub height of the wind turbine.
There are several methods of calculation possible for each main check and the module also includes three supplementary calculation checks:
1. Seismic hazard 2. Lightning rate
3. Extreme and Normal temperature range.
The calculation results of site compliance summarizes the outcome of the seven main checks and three supplementary checks for the wind park and clearly highlights the critical risks that does not comply with the site conditions (EMD International A/S, 2015).
2.5 WIND TURBINE DESIGN CLASS
To assess the technology to be employed at a particular site one needs to adhere to
the specifications set by the International Electro-technical Commission. WindPRO
defines the IEC 61400-1 ed. 3 (2010) specifications of the wind classes using the extreme
50 year gusts, annual mean wind speed at hub height and the turbulence intensity as
outlined in Table 3.
Table 3 Specifications for Wind Turbine Design Class.
Wind Speed Class
Extreme 50-year gust (m/s)
Annual Average wind speed (m/s)
IEC I High wind 50 10
IEC II Medium
wind 42.5 8.5
IEC III Low wind 37.5 7.5
S
Specified by
Manufacturer Specified by Manufacturer Turbulence Class
A B C
0.16 0.14 0.12
2.6 ECONOMIC ANALYSIS
The economic viability of a wind power project includes installation costs, operation and maintenance costs and Average Annual Return (AAR) as per the feed-in tariff per kWh to supply electricity to the grid to determine whether the project is cost effective or not. According to the IRENA Working Paper (2012), the installations costs for offshore wind farms ranges from USD $4000 to USD $4500 per kW and this includes overall costs. Also, operation and maintenance costs are provided as a function of energy production from the wind turbines of offshore wind farms. The operation and maintenance costs ranges from USD $0.027 to USD $0.048 per kWh of electricity generated (IRENA, 2012).
In Fiji, the feed-in tariff for electricity supplied to the national electricity grid from any kind of renewable energy resource by Independent Power Producers is priced at $0.30 FJD/kWh (Fiji Commerce Commission, 2014) which amounts to $0.16 USD/kWh using average exchange rates (1 FJD = 0.5333 USD).
In a simple payback analysis computation the revenue is compared with the costs
and the length of time required for recovering the initial investment costs. The payback
period in years equals to the total capital costs of the wind energy system divided by the annual revenue generated from the energy produced (Manwell et al., 2009). In equation form the simple payback period is expresses as:
𝑆𝑃 = 𝐶
𝑐𝐴𝐴𝑅 = 𝐶
𝑐𝐸
𝑎× 𝑃
𝑒(8)
Where: 𝑆𝑃 is the simple payback period 𝐶
𝑐is the total installation cost 𝐴𝐴𝑅 is the average annual return
𝐸
𝑎is the annual energy production (kWh/year) 𝑃
𝑒is the feed-in tariff for electricity ($/kWh)
The cost of energy (COE) is the unit cost to produce energy in $/kWh from a wind energy system. In the form of an equation it is given as:
𝐶𝑂𝐸 = 𝑇𝑜𝑡𝑎𝑙 𝑐𝑜𝑠𝑡𝑠
𝐸𝑛𝑒𝑟𝑔𝑦 𝑝𝑟𝑜𝑑𝑢𝑐𝑒𝑑 = [(𝐶
𝑐× 𝐹𝐶𝑅) + 𝐶
𝑂&𝑀]
𝐸
𝑎(9)
Where: 𝐶
𝑂&𝑀is the average annual operation and maintenance cost 𝐹𝐶𝑅 is the fixed charge rate
The fixed charge rate refers to the value of interest one pays and or an average annual charge used to account for debt, equity costs and taxes etc.
A study on assessing the impact of renewable technologies on costs and financial
risk of electricity generation in Fiji by Dornan and Jotzo (2011) outlines the cost of
electricity generation in Fiji by different technologies (sources) and this is presented in
Table 4.
Table 4 Present Costs of Electricity Generation in Fiji by Different Technologies.
Generation Type Present Costs (FJD/kWh) Costs Converted to USD/kWh
Hydro-power $0.20 $0.10
Oil-power $0.39 $0.21
Bagasse $0.28 $0.15
Biomass $0.23 $0.12
Onshore Wind-
Power $0.93 $0.49
Studying the cost of electricity generation it can be stated that hydro power,
Bagasse and Biomass are amongst the cheapest sources of energy generation when
compared with oil power and onshore wind power.
CHAPTER 3 SIMULATIONS AND SETUP
WindPRO is a software developed by the Danish Company EMD International and is a very useful tool for wind resource assessment, site suitability assessment, digitalizing information on maps such as contour heights and energy calculations. The calculation for generation of noise and shadows, making photo montages of the landscape with wind turbines, wind turbine technology selection and presentation of results in most of the wind project developing processes can be performed using WindPRO. It has various modules such as Energy and Siting, Environmental, Visualization, Electrical and Economy. A number of modules from the WindPRO software have been utilized for this study.
3.1 SIMULATION PROCEDURE
The WindPRO project for the study was setup by importing the geo-reference map of Bligh Waters Fiji into WindPRO with appropriate country, coordinate and datum information. The WindPRO Meteo object was used to carry out the wind resource assessment of available online wind datasets in and around Bligh Waters. The MCP Module of the WindPRO software was used to perform wind data correlation and prediction of the best wind datasets closest to the micro-site. The wind turbine technology was selected from the WTG catalogue in WindPRO for two model offshore wind farms, one 10 MW and the other 11.5 MW with two different wind turbine types. The Site Compliance module was used to assess the suitability of the WTGs used in the model offshore wind farms with the appropriate wind turbine design class as per the site characteristics. Energy calculations was performed for both the model offshore wind farms using the WindPRO Energy Park module.
3.1.1 WINDPRO PROJECT SETUP
The map of the Blight Waters Fiji to be used into the WindPRO software was
obtained from google earth as a Geo-reference map after three locations on the map were
anchored in terms of their latitudes and longitudes. A new project file was opened into WindPRO software and the project was named and the location of the country was selected to be Fiji. The correct time zone was chosen after which the coordinate system for the site as well as the datum (WGS84) were selected. The geo-reference map was uploaded into WindPRO after defining and correcting the anchored points for the reference of the software with the exact location. The project was ready to work on once the project file with the map was initialized into WindPRO.
3.1.2 WIND DATA
The wind resource data which is mainly the wind speed and the wind direction for the site have been downloaded via the “Go Online” search option of the Meteo object in the WindPRO software online database. It has short and long term atmospheric reanalysis wind data from satellites and meteorological sites all throughout the region. According to WindPRO online database there are various types of climate reanalysis datasets such as:
1. The NCAR (National Center for Atmospheric Research) Basic reanalysis model which is a global data assimilation of a wide range of measured climate data sources model. It has a spatial resolution of 2.5 degrees and the temporal resolution is 6 hours.
2. The QuikSCAT (QSCAT) data is sponsored by National Aeronautics and Space Administration (NASA) Ocean Vector Winds Science Team and is produced by remote sensing microwave scatter meter mounted on the QuikBird satellite. It has a spatial resolution of 0.25 degrees and the temporal resolution is 12.5 hours.
3. METAR – the Aviation Routine Weather Report dataset is based measurements from various airports and permanent weather stations around the globe and at present about 5000 such stations are included in this global dataset. It has a temporal resolution of 1 hour.
4. The Surface Synoptic Observation (SYNOP) data set is created on the
measurements from operated and automated weather stations around the globe.
At present there are about 7000 stations included in this global dataset. It has a temporal resolution of 1 hour and 3 hours.
5. The Modern Era Retrospective-Analysis for Research and Applications (MERRA) data originates from the Global Modeling and Assimilation office of National Aeronautics and Space Administration (NASA). The model grid is 0.5 degrees latitude and 2/3 degree longitude and the temporal resolution is 1 hour.
6. The CFSR-E dataset is an extended version of Climate Forecast System version 1 (CFSR v.1) dataset which was developed by the National Oceanic and Atmospheric Administration (NOAA), National Centers for Environmental Prediction (NCEP) and National Weather Service’s (NWS). The spatial resolution is 0.5 degrees and temporal resolution is 1 hour.
7. The Climate Forecast System Reanalysis (CFSR) dataset is the original reanalysis dataset and was developed with a grid resolution of 0.3 degrees and temporal resolution is 1 hour.
8. The Climate Forecast System version 2 (CFSv2) dataset is a revised version of Climate Forecast System Reanalysis (CFSR) model and has a grid resolution of 0.2 degrees and temporal resolution is 1 hour.
9. Blended Coastal Winds data is the coastal region wind dataset which originates from the USA based National Climate Data Center, National Oceanic and Atmospheric Administration (NOAA) and National Environmental Satellite, Data, and Information Service (NESDIS). The datasets are ocean surface winds and wind stresses on a global 0.25 degrees grid with a temporal resolution of 3 hours.
10. EMD-Global Wind Data (based on ERA-Interim) is a global dataset which is
processed by EMD as a reference wind data for use as a global dataset. The
dataset being a recent global reanalysis dataset from the European Centre for
Medium-Range Weather Forecasts (ECMWF) is derived from the ERA-Interim
dataset. There are a variety of surface parameters in the full ERA-Interim dataset
for weather conditions including the ocean wave and land surface conditions.
The dataset provided by EMD is focused on the main parameters which are relevant for wind-energy purposes. The temporal resolution is 3 hours.
The above mentioned atmospheric reanalysis datasets from the WindPRO online database have been used for the purpose of offshore wind resource assessment of Bligh Waters, Fiji. The locations of datasets where measurements are available are shown in Figure 6 in Bligh Waters and surrounding areas. There are 34 datasets which are distributed around the center of Bligh Waters.
Figure 6 Location of Datasets around Bligh Waters (Source: WindPRO). Plus sign shows the center of Bligh Waters while the colored circles represent datasets.
3.1.3 WIND PREDICTION
The data from the 10 atmospheric reanalysis datasets for the different locations
within and around Bligh Waters have been predicted using the WindPRO MCP Module.
In the MCP module the MCP source data (meteorological measurements) option have been selected as “Use a short-term time series as site and a long term time series as reference.” Each dataset has been correlated with other 33 datasets in pairs of two at one time. One dataset is selected as local measurement (site data) which is usually the short- term data and the other is selected as long term reference data depending on their time series. The full period time series appears in a graph once loaded in the form of two different graphs with the data overlap periods (concurrent data).
Figure 7 Local short-term data and long-term reference data for the best datasets close to the micro-site.
Figure 7 shows the local measurements (short-term) data identified with a blue color while the long-term reference data with a red color. Different options can be selected in terms of wind speed, wind direction and wind energy to be represented in graphic form for both the datasets on averaging of none, one day, one week, one month and one year.
For correlation, WindPRO allows a specified deviation in terms of Maximum
difference in time stamps between local and reference data and refers to them as
concurrent. The concurrent data can be inspected directly as it is represented in the form
of a long table which includes all concurrent data. The wind speeds less than a specified
velocity and difference in wind direction larger than the specified veer are grayed out and not included in the correlation calculations. The standard limits are 4 m/s and 99 degrees.
Two methods of Measure Correlate Predict (MCP) have been used for correlation and prediction which are linear regression method and the matrix method. The linear regression tool allows the user to view the relationship in the form of an animated graph.
Figure 8 Linear Regression Analysis of wind speed for site (short-term) with long term reference.
Figure 8 presents the linear regression analysis of the wind speed measurements at
the site (short-term) with the wind speed measurements of the long term reference. The
linear relation in the plot clearly shows a good correlation in between the site and the long
term reference values. The computed correlation values are outlined next after the long
term prediction of site data.
Figure 9 Regression MCP Wind Speed Prediction of site wind data with long term reference.
Figure 9 presents the wind speed prediction of the site wind speed data with the long term reference data using the Regression MCP method. The correlation coefficient of wind speed is computed to be 0.9894 and 0.9908 for wind direction using monthly averages.
The matrix method represents the model of wind speed and wind direction changes
in the form of a joint distribution fitted on the ‘matrix’ of bins for wind speed and wind
direction. The model permits the user to select the polynomials fitted to the data statistics
or to use measured samples instantaneous when the matrix MCP is done.
Figure 10 Matrix MCP Analysis of wind speed and wind direction at reference position.
Figure 10 presents the Matrix MCP analysis of wind speed and wind direction at reference position with bins of wind speed and wind direction in different wind speed and wind direction ranges.
Figure 11 Matrix MCP Wind Speed Prediction of site wind data with long term reference.
Figure 11 presents the wind speed prediction of the site wind speed data with the long term reference data using the Matrix MCP method. The correlation coefficient of wind speed is computed to be 0.9895 and 0.9914 for wind direction using monthly averages.
For both the methods mentioned, the data for the correlation and the long term wind prediction can be saved and written to a Meteo object. Thus, the best correlation results in the range of 0.80 – 1.00 closest to the micro-site within the Bligh Waters will be written to a Meteo object to represent the wind speed and wind direction of Bligh Waters and for further use in the model wind farm annual energy production calculations.
3.1.4 WIND TURBINE TECHNOLOGY SELECTION
The WindPRO insert New WTG object have been used to select and position the wind turbines used in this study. The Vestas V80-2.0MW offshore wind turbine with a hub-height of 67 meters and the Siemens SWT-2.3-93 2,300 kW wind turbine with a hub height of 68.3 meters have been selected from the built-in WTG catalogue of WindPRO while checking its availability commercially as well. The design standard to be used for the wind turbines have been selected as IEC IA the reason being that Fiji experiences two tropical cyclones on average per year during the wet season which extends from November to April and the maximum wind speed ranges above 100 km/hr (Weir and Kumar, 2008) which is approximately above 28 m/s. These wind turbines will be assessed in terms of energy production and capacity factor in the model 10-12 MW offshore wind farms.
3.1.5 SITE SUITABILITY ANALYSIS
To carry out site suitability analysis of the model 10 MW and 11.5 MW offshore wind farms in Bligh Waters the Site Compliance Module in WindPRO have been used.
The methodology is presented below for the 11.5 MW wind farm.
In the Site Compliance module the analysis is named as Bligh Waters 11.5 MW wind farm followed by selection of the option to be used for the site and layout checks which is Mast data and flow model. Both flow models are used, WEng 3 (WAsP Engineering) and WAsP using long-term corrected wind statistics and then the design standard: IEC 61400-1 ed. 3 (2010) for the specific selection is made as IA for the WTG design class.
The mast data to be used is selected. Here a number of wind datasets can be selected to be used for calculation and the Main height must also be selected. For the calculation of vertical wind shear, the wind speed at multiple heights is selected from one of the datasets closest to the micro-site which has wind speeds available at multiple heights of 10 m, 25 m, 50 m, 75 m, 100 m, 150 m and 200 m.
The wind turbines are selected and the specific mast data to be used is selected either in terms of nearest mast or by manual selection of a particular mast dataset of choice.
For this study the nearest mast data is selected which is the long-term predicted wind data from the MCP analysis.
The site data is selected for WAsP calculations and the WAsP calculation is performed and then there is a tick in front of the WAsP header to indicate proper execution.
For the WEng calculations the site data is selected together with the “Advanced”
option for reduced geostrophic wind and the selection of the recommended Turbulence calculation model as Kaimal and hence the WEng calculation is performed and then there is a tick in front of the WEng tab header to indicate proper execution.
Finally, in the calculation module all the checks to be included can be selected and
then the calculation is executed. After the completion of the calculation execution the
results can be viewed for the different checks in terms of three different colors indicating
results as:
“Ok with green color” meaning “No WTGs exceed IEC limits”
“Caution with orange color” meaning “≥ 1 WTG exceed IEC limits – exceedance not considered critical”
“Critical with red color” meaning “≥ 1 WTG exceed IEC limits – exceedance potentially critical.”
3.1.6 WIND FARM ENERGY CALCULATION
The WindPRO Energy PARK Module has been used to perform the annual energy production computation from the best correlated results of the atmospheric reanalysis datasets closest to the micro-site. The PARK calculation of the Energy Module have be used to perform the AEP calculations and wake losses for the model 10 MW and 11.5 MW offshore wind farms in Bligh Waters, Fiji using WindPRO. The inputs for the PARK calculation are the positions of the WTGs, type of WTG, and the hub-height of the WTGs and the best correlated wind data from the Meteo object created earlier in MCP. The options in the main Park Module were AEP calculations, model parameters – terrain type:
Offshore & Water areas and the Wake model being N. O. Jensen (RISØ/EMD) and result – 10 % to account for uncertainties in the data and calculation modules. Also, the wind shear used for the calculations have been calculated and used as 0.10.
3.2 ECONOMIC ANALYSIS
The cost of investment and the operation and maintenance costs have been adopted
from literature (IRENA Working Paper, 2012) and maximum costs have been used
because Fiji is new to offshore wind so related costs will be high and also as Fiji
experiences on average two tropical cyclones per year so maximum operation and
maintenance costs have been used. The procedure to calculate the simple payback period
and the cost of energy generated is also adopted from literature (Manwell et al. 2009). The
cost of energy generation by the model offshore wind farms is compared with the current
electricity generation costs from literature (Dornan and Jotzo, 2011) to outline the
feasibility of offshore wind to support the national electricity grid in Fiji. The fixed charge rate is taken to be an average of 5.69 % (Whiteside, 2014) taking into account that renewable energy projects have a tax holiday of 5 years in Fiji and with the benefit of importing renewable energy equipment with zero percent fiscal tax (Fiji Revenue and Customs Authority, 2013).
The Fiji governments interests, efforts and focus towards generating electricity from renewable energy resources has given rise to international banks such as the Australia New Zealand Banking Corporation (ANZ), Westpac Banking Corporation and local banks such as the Fiji Development Bank (FDB) and the Reserve Bank of Fiji (RBF) to actively involve themselves into microfinance schemes to financing renewable energy projects for rural electrification as well as bigger feasible independent projects at lower interest rates (Pacific Islands Trade and Invest, 2011).
Therefore, the methodology adopted and utilized for sections 3.1.1 - 3.1.6 for this
study have been summarized above and for detailed explanations about the different
sections it can be referred to from the WindPRO 2.9 User Manual (2013).
CHAPTER 4 RESULTS AND DISCUSSION
Simulations have been performed as per the procedure outlined in chapter 3 and results are presented and discussed here.
Offshore wind resources of Bligh Waters in Fiji have been assessed using wind resources data from the atmospheric reanalysis datasets available from WindPRO online database for Bligh Waters. Correlation and prediction of wind data were performed for the 34 datasets using the MCP (Measure Correlate Predict – long term correction – STATGEN) module of the WindPRO software using the methods of Linear Regression MCP and Matrix Method MCP. The details about the 34 datasets can be viewed in the appendix (Tables 10 and 11).
Table 5 Best Correlation Results of Wind Speed and Wind Direction using Linear Regression
and Matrix MCP Methods.
Table 5 presents the correlation results with quality of reference as “Good (0.8 – 0.9)” and “Very Good (0.9 – 1.0)” according to WindPRO 2.9 User Manual (2013) for MCP analysis of the datasets for wind speed and wind direction using the linear regression and the matrix method of the MCP module of the WindPRO software. The correlation results of 10 pairs of datasets ranges from 0.8665 to 0.9817 for wind speed and 0.9882 to 0.9989 for wind direction using the linear regression MCP and the matrix MCP.
Correlation values are very similar for both the methods used. Since the long term datasets are for 30 years therefore, it provides a better representation of wind speed and wind direction at Bligh Waters.
The best correlated wind speed closest to the micro-site are given by CFSR2 E178.363 S17.274 and CFSR-E E179.00 S17.00 datasets and using these datasets the mean wind speed is computed to be 6.51 m/s at a height of 10 m. The mean wind direction is 121.2°, the Weibull mean is 6.51 m/s and the Weibull A parameter which is used to indicate on average how windy the site is 7.35 m/s and the Weibull shape parameter k, which outlines how peaked the wind distribution is 2.2981 and these values correspond well with the research done in Fiji from literature (Kumar and Prasad, 2010; Singh, 2015).
Figure 12 Radar Diagram of the Mean Wind Speed and Frequency at Bligh Waters in Fiji.
Figure 12 presents the mean wind speed in the form of a radar diagram which provides the mean wind speeds in the different direction sectors and also outlines the frequency of occurrence with respect to the wind direction and this shows the dominant wind direction as East-southeast (ESE) and South-southeast (SSE).
Moreover, the plot of the wind roses for most the wind resources assessed within and around Bligh Waters in Fiji is shown in Figure 13.
Figure 13 Plot of Wind Roses for most of the Wind Resources Assessed showing Dominant Wind Direction (Source: Original Map – Google Earth).
Studying the plots of the wind roses from Figure 13, it can be explicitly understood
that the dominant wind direction of Bligh Waters and its surrounding region is East-
southeast (ESE) and South-southeast (SSE) which corresponds well with the research
done in Fiji from literature (Kumar and Nair, 2013 & 2014; Singh, 2015). Dataset 7 is the
closest to the micro-site.
Using the wind speed data from EmdERA E178.562 S17.193 which is available at different heights ranging from 10 to 200 meters and then calculating the wind shear coefficients using the power law in the form of equation 3 are expressed in Table 6.
Table 6 Wind Shear Coefficients using Mean Wind Speed from EmdERA E178.562 S17.193 at Different Heights.
Height (m) Wind Speed (m/s) Shear Coefficient
10 5.35 0.20
25 5.77 0.08
50 6.08 0.08
75 6.27 0.08
100 6.40 0.08
150 6.59 0.08
200 6.72 0.08
Average 0.10
Looking at Bligh Waters, Fiji and referring to Table 1 (Patel, 1999) of wind shear coefficients of various terrains it can be stated that Bligh Waters falls in the terrain type of ‘Lake, ocean, and smooth-hard ground’ and thus, it will have a wind shear coefficient of 0.10. This is also checked and confirmed from Table 6 which shows the wind shear coefficient calculated at various heights. It can be noted that the wind shear coefficient calculated ranges from 0.08 to 0.20 and the average wind shear coefficient being 0.10 which corresponds to the terrain type with ‘Lake, ocean, and smooth-hard ground’ using Table 1 (Patel, 1999). The calculated average wind shear has also been used for the energy calculations.
The diurnal pattern of mean wind speed and mean wind direction of the best
correlated datasets at the micro-site in Bligh Waters Fiji are presented in Figures 14 and
15 respectively.
Figure 14 Diurnal Mean Wind Speed of Bligh Waters in Fiji.
Figure 15 Diurnal Mean Wind Direction of Bligh Waters in Fiji.
The values of the diurnal mean wind speed at Bligh Waters Fiji ranges from 6.06 m/s to 6.75 m/s and diurnal mean wind direction ranging from 116.8 to 129.1° at a height of 10 meters respectively during a 24 hour period.
6 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Wind Sped (m/s)
Hour
Diurnal Mean Wind Speed
116 118 120 122 124 126 128 130
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Wi nd D ir ec ti on (° )
Hour
Diurnal Mean Wind Direction
Similarly, the annual pattern of mean wind speed and mean wind direction of the best correlated datasets to represent the annual wind speed and wind direction at the micro- site in Bligh Waters Fiji are presented in Figures 16 and 17 respectively.
Figure 16 Annual Mean Wind Direction of Bligh Waters in Fiji.
Figure 17 Annual Mean Wind Direction of Bligh Waters in Fiji.
The values of annual mean wind speed at Bligh Waters ranges from 5.43 m/s to 7.40 m/s and annual mean wind direction ranging from 103.7 to 126.5° at a height of 10
5 5.2 5.4 5.6 5.8 6 6.2 6.4 6.6 6.87 7.2 7.4 7.6
1 2 3 4 5 6 7 8 9 10 11 12
Mean Wind Speed (m/s)
Month Annual Mean Wind Speed
100 102 104 106 108 110 112114 116 118 120 122 124 126 128
1 2 3 4 5 6 7 8 9 10 11 12
Wind Direction (°)
Month Annual Mean Wind Direction
meters during a year’s period. Both the diurnal and the annual mean wind speeds correspond well with the studies done by Kumar and Prasad (2010) as well as Singh (2015).
The wind turbine technology selected for Bligh Waters Fiji is presented in Table 7.
Table 7 Wind Turbine Technology Selected for Bligh Waters, Fiji.
General
Information Turbine Name SWT-2.3-93
V80-2.0MW offshore 2000
Manufacturer Siemens Vestas
Commercial
Availability Available Available
Operating Data
Rated Power 2,300 kW 2,000 kW
Cut-In Wind Speed 4 m/s 4 m/s
Rated Wind Speed 13-14 m/s 16 m/s
Cut-Out Wind Speed 25 m/s 25 m/s
Maximum 3 s Gusts 59.5 m/s -
Wind Class IEC IA IEC IA
Rotor Diameter 93 m 80 m
Swept Area 6,800 m
25,027 m
2Power Density 2.95 m
2/kW 2.51 m
2/kW
Operational Interval 6.0-16 rpm 10.8-19.1 rpm Power
Regulation Type
Pitch Regulated with Variable Speed
Pitch Regulated with Variable Speed Electrical
Data Generator Type Asynchronous
4-pole asynchronous with variable speed Generator Power 2,300 kW/unit 2,000 kW/unit
Speed 1500 rpm -
Output Voltage 690 V 690 V
Frequency 50/60 Hz 50/60 Hz
Tower Hub Height 68.3 m 67.0 m
The wind turbine technology for Bligh Waters, Fiji have been selected to be Vestas
V80-2.0MW offshore wind turbine with a hub height of 67 metres and Siemens SWT-2.3-
93 2,300 kW with a hub height of 68.3 meters and the wind turbine class as IEC IA for
both wind turbines as per the site compliance module analysis in WindPRO. Also since Vestas and Siemens wind turbines are amongst the best offshore wind turbines to be used for offshore applications.
The site suitability analysis of the model 10 MW wind farm using five Vestas V80- 2.0MW offshore wind turbines and 11.5 MW wind farm using five Siemens SWT-2.3-93 2,300 kW wind turbines in Bligh Waters Fiji are presented in Figures 18 and 19 respectively.
Figure 18 Site Compliance Results of the 10 MW model Wind Farm using Vestas Wind Turbines
in Bligh Waters Fiji.
Figure 19 Site Compliance Results of the 11.5 MW model Wind Farm using Siemens Wind Turbines in Bligh Waters Fiji.
According to the analysis done by the site compliance module in WindPRO for both the wind turbine types (Vestas and Siemens), it can be reported that for both wind turbine types out of the seven main checks two checks that is wind distribution and wind shear have a result of “Caution” while the others checks have the result as “OK”. And for the other three supplementary checks only one check that is seismic hazard is “Caution”
while the remaining two are “OK”. A result of “OK” means “No WTGs exceed IEC
limits” while a result “Caution” means “≥ 1 WTG exceed IEC limits – exceedance not
considered critical.” WAsP and WAsP Engineering applications in the WindPRO software have been used to carry out this analysis.
The energy production from the model 10 MW wind farm and the 11.5 MW wind farms using different wind turbines with the best correlated results from the linear regression MCP close to the micro-site in Bligh Waters is presented in Figures 20 and 21 respectively. The energy calculation have been done using the WindPRO Energy module – Park (Wind farm AEP based on MODEL or METEO) at the hub height of 67 metres and 68.3 meters respectively using a wind shear of 0.10 which represents offshore and water areas.
Figure 20 Calculated Annual Energy from the 10 MW Model Wind Farm in Bligh Waters Fiji.
It can be reported that the 10 MW model offshore wind farm produces 30,909.2
MWh energy annually with a capacity factor of 35.3 %, park efficiency of 99.8 % and full
load hours of 3091 hours/year. Each Vestas V80-2.0MW offshore wind turbine produces
an average of 6,181.8 MWh annually at a mean wind speed of 7.83 m/s.
Figure 21 Calculated Annual Energy from the 11.5 Model Wind Farm in Bligh Waters Fiji.
The 11.5 MW model offshore wind farm produces 40,327.5 MWh energy annually with a capacity factor of 40.0 %, park efficiency of 99.8 % and full load hours of 3507 hours/year. Each Siemens SWT-2.3-93 2,300 kW wind turbine produces an average of 8065.5 MWh annually at a mean wind speed of 7.84 m/s.
The calculated Annual Energy Production (AEP) of the wind farms is a 10 %
reduced value from the actual calculation to account for errors in wind data, correlation
calculations, power curve and losses due to wake interaction (WindPRO 2.9 User Manual,
2013). This energy calculation cannot be directly compared to the existing studies done in
Fiji as those done so far are using few small scale wind turbines. But comparing the AEP
of the model offshore wind farms with the existing onshore wind farm it can be reported
that the offshore wind resources have a higher potential compared to the onshore wind
resources. The 10 MW onshore wind farm has an AEP of 5,348 MWh for the year 2013
(FEA Annual Report, 2013) while the model 10 MW and 11.5 MW offshore wind farms
have an AEP of 30,909.2 MWh/year and 40,327.5 MWh/year respectively. There is a huge
difference because the existing onshore wind farm has also been poorly planned (Fiji Times Online, 2009) with lower hub-height and smaller 2 bladed wind turbines and also because offshore wind resources are much higher in comparison with onshore wind resources.
The economic analysis of the 10 MW and the 11.5 MW model offshore wind farms using Vestas and Siemens wind turbines in Bligh Waters Fiji are presented in Tables 8 and 9 in terms of cost of installation, operation and maintenance costs, annual average return, simple payback period (SP) and cost of energy (COE).
Table 8 Economic Analysis of the 10 MW model wind farm using Vestas Wind Turbines in Bligh Waters, Fiji.
Installed
Capacity Cost per MW O&M Costs per kWh
AEP in kWh/year
Feed-in
tariff / kWh FCR
10 MW USD $4,500,000 USD $0.048 30,909,200 USD $0.16 5.69%
Equation used for Calculation Calculation Result
Capital Costs = Installed Capacity x Cost/MW 10 x $4,500,000 $45,000,000
AAR = Feed-in tariff x AEP $0.16 x 30,909,200 $4,945,472
O&M Costs = O&M costs x AEP $0.048 x 30,909,200 $1,483,642
Simple Payback Period = Capital Costs / AAR $45,000,000 / $4,945,472 9.1 years Cost of Energy = [(Capital Costs x FCR) +
O&M Costs] / AEP
($45,000,000 x 0.0569) + $1,483,642]
/ 30,909,200 $0.13
Performing a simple economic analysis of the 10 MW model offshore wind farm
in Bligh Waters Fiji it can be reported that the cost of installation is USD $45,000,000 and
average annual return per annum is USD $4,945,472, operational and maintenance cost
per annum is USD $1,483,642 and hence, the wind farm has a payback period of
approximately 9 years and the cost of unit energy generation is computed to be USD $0.13.
Table 9 Economic Analysis of the 11.5 MW model wind farm using Siemens Wind Turbines in Bligh Waters, Fiji.
Installed
Capacity Cost per MW O&M Costs per kWh
AEP in kWh/year
Feed-in
tariff / kWh FCR
11.5 MW USD $4,500,000 USD $0.048 40,327,500 USD $0.16 5.69%
Equation used for Calculation Calculation Result
Capital Costs = Installed Capacity x Cost/MW 11.5 x $4,500,000 $51,750,000
AAR = Feed-in tariff x AEP $0.16 x 40,327,500 $6,452,400
O&M Costs = O&M costs x AEP $0.048 x 40,327,500 $1,935,720
Simple Payback Period = Capital Costs / AAR $51,750,000 / $6,452,400 8.0 years Cost of Energy = [(Capital Costs x FCR) +
O&M Costs] / AEP
($51,750,000 x 0.0569) + $1,935,720]
/ 40,327,500 $0.12
