The Wind Power Potentials In Cyclades Environmental Sciences Essay

Abstract

The current paper presents an investigation of wind power potential in Cyclades, Greece and to provide the reader with the essential information and considerations of siting a wind farm in the specified location. Three representative Islands have been chosen in order to fulfil our task of quantifying the wind power potentials in Cyclades as well as comparing the performance of a particular wind turbine. Analytically on the three locations, four wind turbines from different manufactures have been chosen with same rotor diameter to compare amongst them. Furthermore the data analysis will be performed by using Rayleigh distribution in order to quantify the actual power output supplied from each wind turbine during one year. As soon as the rating of the wind turbines will be found, full siting process will be analytically explained and demonstrated involving all the considerations that have to be made for ensuring the feasibility of a wind farm.

Introduction

The theoretical wind power availability could potentially cover Europe's total need for energy and even generate a surplus for future needs and export. Wind energy is one of the most developed renewable technologies in Europe due to the nature and the availability of the source (wind). Each EU member state has a set of national objectives in terms of the percentage of its gross national electricity consumption that is being produced from renewable sources. All the member states have their individual targets as they are being specified in Directive 2001/77/EC [1] . Within Europe, Greece has great potential of renewable energy development especially wind energy. In contrast with Greece's rich natural wind source, Greece has stayed behind the rest of the member states mainly due to lack of infrastructures, an old bureaucratic system for planning approvals which has been updated recently and its specific topography. However, the need for expansion on the sector and high wind availability has opened great opportunities and potentials making Greece and especially Aegean and Ionian Sea ideal locations for placing wind farms. In this Study we shall concentrate on Cyclades that are being characterised from high wind source that has not been exploited yet. Most of the Cycladic Islands are not connected with the National Grid and they use small diesel stations to cover their energy requirements. Cyclades is a group of more than two hundred islands (approximately 220) located east of Athens and south Aegean Sea. The topography of Cyclades is being characterised from high hills and mountains. The climate is basically Mediterranean (temperate) resulting in mild wet winters and hot-dry summers.

History

Humans have been using wind power at least since 3,000 BC as propulsion system for their boats. First in the area of Eastern Mediterranean sea the Ancient Egyptians as well as Greek and Semitic tribes were using sails for their ships and the Babylonian civilization under the reign of Emperor Hammurabi was the first to use wind for the great project of irrigating the Tigris and Euphrates valley. However the first man who actually extracted power from wind was the Famous Greek engineer, Heron of Alexandria who developed the first wind driven wheel to power a machine in the 1st century AD [2] . In the next years and until the 1950s only minor improvements and developments have been done on the field caused from the cheap labour in the ancient and colonial ages and the steam engine in the industrial revolution. Nowadays, the oil price has reached really high levels resulting in developments on the field on a year by year basis.

Literature review

Prior placing the wind turbines, the optimum location needs to be obtained. A particular wind turbine model can produce quite different power output depending on the location. In order to cover all the aspects of wind farm installation a siting process has been developed. The siting process of a wind farm can be broken down into the following steps according to Hiestel and Pennell, 1981:

Identification of geographic areas needing further study.

Selection of candidate places.

Preliminary evaluation of candidate site.

Final site evaluation.

Micrositing.

Each stage aforementioned includes a series of actions that have to be taken in order to proceed with the installation of the wind farm. In the first stage, promising areas within the desired location have to be identified. Wind atlases can be used as guide in order to start filtering the desired location in terms of wind availability. Wind atlases give an indication of the mean wind speed, but these data cannot be used for implementing the power density calculations and in order to proceed proper measurements have to be taken. In the second stage and as soon as we have found the promising locations for placing wind farms, we consider if it is feasible to install wind turbines in the specified locations. Topographical considerations and engineering problems such as transportation and erection of the wind turbines are taken into account as well as an indication of public acceptance and behaviour towards the project is required. This is a very important stage due to the fact that most wind farm proposals are being rejected for non compliance with the location specifications. In the third stage, all the candidate sites have potentials for locating wind farm. However this stage includes ranking of the candidate sites in order to converge into a solution which fulfils the following aspects:

Environmental impact.

Public acceptance.

Operational problems.

Economical potential.

Compared with the previous phase in this stage in-depth studies are being carried out in order to make sure that the aforementioned aspects are being met. In the fourth stage, measurements should include all the parameters which affect the wind turbine's output such as wind shear and turbulence. This is experimental phase and process of the data collected is essential. It is important to collect accurate data because it would be difficult to obtain any errors during Micrositing stage as many consultant companies insert those data into computer software. In order to implement Micrositing stage, numerous codes which integrate and process the input data resulting in high quality solutions. It is worth saying that no solution is 'correct', there must be a balance between the advantages and disadvantages that each choice is related. However, at the end it should be the human element which has to evaluate and choose the proposals produced from the applied software.

In order to indicate the wind energy potential five parameters have been set according to World energy council (1993). The optimum location for the wind farm should fulfil all the criteria set on each section. Even though that these parameters could be considered as siting process, they lack many aspects from the formal siting process aforementioned and ideally should be part of the 'Identification of geographic areas needing further study' or 'Selection of candidate places' phases. Analytically the parameters are as follows:

Meteorological potential. After more than twenty year of existence of wind power, this parameter is not considered as unknown, as numerous studies have tried to quantify the available wind source. At the moment globally we have estimated the available wind source in terms of mean wind speeds that are enough to distinguish between promising places and inappropriate locations. However further wind measurements have to be carried out in order to fully assess the location in terms of wind availability.

Site potential. As soon as we have spotted promising areas within the desired location, we have to filter our search with taking into account geographical availability. For instance in an area where the mean wind speed velocity is 9 m/s i.e. highly promising, in some locations within that particular area, it may not be appropriate for placing wind turbines. This could happen due to unsuitability of the soil's quality in order to place the foundations (it may be proved expensive to establish foundations in that particular area), inclination of the ground for installing the wind turbines safely or any topographical implication in general.

Technical potential. Practically the technical potential is the combination of the site potential with the available technology at present time. Indirectly technical potential is related with meteorological potential due to the relation of meteorological potential and filtered meteorological potential (site potential). This series of parameters can be characterised as cascade as the previous parameter affects and is being added to the next one.

Economic potential. With having an idea of technical potential, it should be realised in economical terms. The requirements in order to assess the economic potential are the cost of technology, transport and installation. Moreover the turbine's life cycle should be included in the analysis as, wind turbines with long life cycle give greater economical opportunities and reduce the risk of investment.

Implementation potential. This final criterion takes into consideration the risks and opportunities of the project as a whole existence. Data about the parameters related with the 'selection of candidate site' are included but not to great detail as the 'preliminary evaluation' that covers relatively in dept, follows.

In Greece, the most frequent reason for failing to install wind farm is the opposition from the local community which is being realised strongly and actively. This opposition in many cases has led to extra cost for the developer's party or in the worst scenario total cancellation of the project. Dimopoulos A. and Kontoleon A. (2009) have tried to analyse the factors that determine the local community's acceptance or rejection of developing new wind farms. This study has launched a survey in two Aegean islands; Naxos and Skyros that took place on the summer of 2007. The authors have included many factors such as: wind farm size, wind turbine height, the annual compensation that each household takes and impact of the installation to tourism. The output of this study was that the benefits of not installing wind farms in protected-sensitive areas should prevail over not only the operation and installation cost but also the peripheral environment. However in Naxos and Skyros where the protected areas are limited, the local population preferred to have wind turbines with a small regulation in height provided that compensation is granted. Furthermore this study has came into compliance with previous studies performed across Europe stating that the developers should try to cooperate with the local authorities as it is regarded highly from the local population.

Data analysis

In order to quantify the wind source we must go according to the following methodology:

Obtain the variation of wind velocities in m/s and how many hours these velocities are blowing throughout the year. Determine the mean annual velocity for each site.

Determine the probability density function using Rayleigh and Weibull distribution.

Correction of the air density, as at hub height is different from the sea level.

Correction of wind speed according to logarithmic law as at hub height is different from the level measured (anemometer height).

Weather stations

Wind velocities have been obtained from three locations across Cyclades. The following table (table 1) summarises the details of the chosen locations. These data have been granted from Centre for Renewable Energy Sources (CRES) based in Athens.

Mean annual velocity

The raw data for the three locations can be shown in the appendix. It is worth to mention that these values from 1 to 25 are not discrete values but ranges. Id est, that the value "x" of a wind speed velocity used measured in m/s is described from. The following formula expresses the relative probability curve defined from the data obtained.

(1)

Prob(U) is the probability density function of wind speed. Practically represents how frequently a particular wind speed occurs.

The data obtained from CRES, for academic purposes the data given are in the following form:

(2)

Equation (2) simplifies the velocity probability distribution approach and makes applicable the use of equation (3) after-mentioned.

(3)

Represents the relative probability of each wind speed range.

Represents the time of occurrence of each wind speed range.

At this point we are able to obtain the mean wind speed velocity from the following equation:

(4)

Therefore for Andros, Naxos and Milos the annual mean wind speeds are 9.008 m/s, 7.678 m/s and 7.783 m/s respectively. The sites could be characterised as Wind Power Class 7 as the annual mean speed is between 7 m/s and 9.4m/s at sea level.

Probability Density Function (p.d.f.)

From bibliography, there are two probability distributions widely used; Weibull and Rayleigh. The difference between the two statistical approaches is the Rayleigh distribution makes use of just the mean wind speed obtained above whereas Weibull distribution makes use of shape factor (k) and scale factor (c) making Weibull distribution more accurate. However for most sites Rayleigh distribution is considered adequate.

Weibull distribution

The variation of wind speeds are best described from Weibull p.d.f. that is a mathematical idealisation of the distribution of wind speeds throughout time. The shape parameter (k), indicates the way that wind speeds are distributed whereas the scaling factor (c) indicates how windy the site is. The Weibull p.d.f. is given from following equation:

Is the wind speed velocity.

Is the shape factor.

Is the scaling factor.

It is not a straight forward process to obtain k and c. There are various methods for determining these parameters. In this paper, k will be obtained by using the following formula (Justus, 1978):

is the standard deviation.

Parameter c is therefore given from the following formula (Lysen, 1983):

For the three sites, the following table summarises the k and c parameters.

Rayleigh distribution

This approach assumes standard wind speed parameters and the only parameter that is used is the mean wind speed velocity. Rayleigh p.d.f. is a standard form of Weibull p.d.f. assuming that the site has winds that are mostly strong (more than 7 m/s) with periods of low and really high wind speeds. For Cyclades this is an adequate approximation that throughout the year gives quite accurate results in the range of 4-5% less in energy production. However if this approximation is being done in a monthly basis the error is considerably higher. The Rayleigh p.d.f. is given from the following formula:

Comparison of Rayleigh and Weibull p.d.f.

Each probability density function indicates the fraction of time that a particular wind speed magnitude predominates. Figures 1-3 illustrate the

Correction of air density at hub height

As temperature increases, the air density decreases resulting in air that is less dense during summer than in winter. This variation is around ten to twenty percent from summer to winter. However by making the analysis in an annual basis this variation does not impose any change to our calculations.

Air density changes also with amplitude. In our analysis we will assume that the temperature of the chosen location is constant throughout the year i.e. for Andros the annual mean temperature is 18.7, for Naxos 18.92 and for Milos 18.2.

Therefore at sea level, the air density for each place can be calculated from:

(3)

Application of the equation (3) yields the air density for Andros, Naxos and Milos to be 1.2095, 1.2086 and 1.2116 respectively. Those air densities will be used in order to determine the effect of amplitude upon air density.

In order to determine the air density at increased amplitude, we should obtain first the atmospheric pressure as the amplitude increases. The following formula yields the atmospheric pressure relative to amplitude:

(4)

represents the atmospheric pressure at sea level

represents the amplitude measured in meters

Represent the gravitational acceleration, gas constant and temperature respectively.

The corrected final density relative to amplitude can be given from equation (5):

(5)

An empirical method for calculating air density relative to amplitude and temperature is represented from the following equation:

(6)

Represents the normal lapse rate which is equal to 6.5 â„ƒ per 1,000 metres increase in amplitude obtained from Paul Gipe (2004).

Therefore the following tables summarise for each location the effect of amplitude in the air density using the two methods. The first method is considered the most accurate and it is based on the assumption that air behaves as ideal gas that in Greece with the dry weather, it is the case. From the analysis, it can be obtained that the variation between the two approaches is not substantial. However for the forthcoming calculation the results from the 1st method will be used. If the aforementioned calculations have not been operated, the error on the analysis will vary from 4-9% depending of the mean annual temperature and amplitude of the installation.

Correction of wind speed at hub height

The wind speed has been measured on Andros, Naxos and Milos at elevation of 12, 9.8 and 10 meters respectively as mentioned on Table 1. However as the elevation increases, so does the wind velocity. This figure does not increase infinitely but can be modelled with use of the power law. Power law can be described from the following formula:

(6)

Represents the wind speed at amplitude Z, is the reference wind speed at amplitude and is the power law exponent. Accuracy in calculations depends strongly upon accurate determination of. Alpha varies with such parameters as amplitude, climate, wind direction and magnitude and type of the terrain. Counihan (1975) proposed the following formula for obtaining alpha in terms of surface roughness:

(7)

Represents the surface roughness of a particular terrain measured in millimetres.

In Cyclades the terrain outside farming and residential areas can be described between Rough pasture and fallow field with surface roughness value of 15 mm. Therefore equation (7) yields a value of equal to 0.37503.Figure 4 visualises the chosen surface roughness.

Figure [3]

Energy production estimate

Four wind turbines will be compared in terms of power production and general performance on the following sites. The chosen turbines will have same rotor diameter and will be from different manufacturers. Due to lack of infrastructure in the Cyclades, the maximum disc diameter of the chosen wind turbines will be 52 m. Therefore the area defined from the rotor disc will be taken as 2,124. Firstly the optimum amplitude for each location will be determined in order to place the wind turbines. The energy in MWhs of each wind turbine is given from the following formula:

Is the Weibull distribution function. It has been prevailed over Rayleigh p.d.f. due to higher accuracy, as it uses more parameters to describe the wind power distribution.

Vestas V52 wind turbine

In the site it is claimed that this turbine produces 2.3, 3, 3.6 GWh for 7, 8, 9 m/s

Gamesa G52 wind turbine

Suzlon S52 wind turbine

Americas Wind Energy AWE52 wind turbine

In order to make the analysis more realistic, the power output required will be set in terms of GWh. This choice is giving ground for the next phase of the project which is wind farm siting.

Siting the wind farm

In this chapter we will deal with the array of wind farm. Furthermore, road construction within the boundaries of the wind farm in order to support maintenance and other activities will be demonstrated. Placing auxiliary systems such as control room is part of this chapter. The following aspects will be covered in detail as well.

Site preparation: Clearing the land, taking topologic measurements about the quality of the soil that the wind turbine will be placed, are two examples of preliminary work prior installation. The process of making site preparation will be included in this part.

Turbine transportation: Transportation is an important parameter especially as Cyclades lack infrastructures (medium size ports, narrow and steep grade roads). This topic is related with the wind turbine model as a big turbine will not be easy to transport on site.

Turbine assembly: This topic covers the assembly and erection of wind turbine. Cranes will be used and the size of the turbine will determine the ease of this process.

Additionally aspects such as:

Environmental impact: this aspect is essential to be minimised as the main purpose of wind farms is to co-exist with the local environment and at the same time produce 'clean' power output. Therefore environmentally sensitive areas, bird flyways and electromagnetic interference will be key parameters that would be taken as criteria for the candidate places.

Public acceptance: In order to secure public acceptance, considerations such as visual impact are essential. In order to minimise the effect of visual impact the wind turbine must be as far as possible from residential and archaeological sites. Cultural issues will be included in this phase.

Land use: Although wind farms are land intensive, they actually require only 2% from the land allocated. However the surrounding land cannot be used for any activity.

Property values: This aspect is related primary with the visual impact that wind turbines have to the surrounding area. It should be noted that generally wind turbines do not tend to cause any reduction of land value but this consideration still has to be include in the siting process.

Noise: Noise propagation, has a considerable impact to the surrounding area. However according to legislation wind farms must be at least 300m away from residential areas. This distance in conjunction with noise regulators mounted on the wind turbine system has showed that wind farms do not mean dense acoustic nuisance.

Public safety: it is important to secure safety within wind farms as the tall towers, control rooms and high voltage areas impose extra care to be taken.

Discussions

In this chapter discussion about the calculations will be made. At this stage we should be able to recommend the optimum wind turbine for each site and comment of the wind farm arrangement.

Just the relative probability and thus mean annual speed is not enough for extracting a clear view of the available wind potential

Conclusions

In this chapter, recommendations for future work as well conclusions will be done.

References

Quaschning Volker.2005. Understanding renewable energy systems. London: Earthscan.

Gipe Paul. 2004. Wind power. London: James & James.

Manwell JF, McGowan JG, Rogers AL. Wind power explained: theory, design and application. 2nd edition. Chichester: Wiley

Papathanassiou Stavros, Boulaxis Nikos. 2006. Power limitations and energy yield evaluation for wind farms operating in island systems. Renewable enegy, 31, pp. 457-479.

Fyrippis I, Axaopoulos P, Panayiotou G. 2010. Wind energy potential assessment in Naxos Island, Greece. Applied energy, 87,pp. 577-586.

Vogiatzis N, Kotti K, Spanomitsios , Stoukides M. 2004. Analysis of wind potential and characteristics in north Aegean, Greece. Renewable energy, 29, pp. 1193-1208.

Kaldellis JK. 2002. Optimum autonomous wind-power system sizing for remote consumers, using land-term wind speed data. Applied energy, 71, pp. 215-233.

Lalas DP, Tselepidaki H, Theoharatos G. 1983. An analysis of wind power potential in Greece. Solar energy, 30, pp. 497-505.

World energy council. 1993. Renewable energy resources: opportunities and constraints 1990-2020. London. World energy council publications.

Dimitropoulos A, Kontoleon A. 2009. Assessing the determinants of local acceptability of wind-farm investment: a choice experiment in the greek Aegean islands. Energy policy, 37, pp. 1842-1854.

Oikonomou E, Kilias V, et al. 2009. Renewable energy sources (RES) projects and their barriers on a regional scale: The study case of wind parks in the Dodecanese Islands, Greece. Energy policy, 37, pp. 4874-4883.

Counihan J. 1975. Adiabatic atmospheric boundary layers: a review and analysis of data collected from period 1880-1972. Atmospheric environment, 9, pp. 871-905.

Justus CG. 1978. Winds and wind system performance. Philadelphia. Franklin Institute press.

Lysen EH. 1983. Introduction to wind energy. Amersfoot. SWD publication SWD 82-1.

Bibliography