Available online at www.sciencedirect.com

ScienceDirect

Energy Procedia 36 (2013) 179 - 188

The Analysis of Wind Power Potential in Sahara Site of Algeria-an Estimation Using the 'Weibull' Density Function

M. DAHBIab;* A. BENATIALLAHb ; M.SELLAM c

a Department of Science and technology,University of Bechar, Algeria, Email: dahbimaamar@yahoo.fr.

b LEESI, Laboratory, University of Adrar, Algeria, Email: benetiallah.ali@gmail.com , cENERGARID, Laboratory, University of Bechar, Algeria Email: Selmab09@gmail.com

Abstract

The wind energy is one of the most significant and rapidly developing renewable energy sources in the world and it provides a clean energy resource, which is a promising alternative in the short term in Algeria. The main purpose of this paper is to discuss the wind potential in Sahara site of Algeria and to perform an investigation on the wind power potential of desert of Algeria. The Weibull density function has been used to estimate the monthly power wind density and to determine the characteristics of monthly parameters of Weibull. A simulation model has been established to describe the characteristics of a particular wind turbine. The monthly power generated and the monthly operating hours by the wind turbine to be simulated. The simulation based on the data wind speed measured at the 10th meters and collected during 8760hours by a wind observation station web site weather underground (The global weather data could be obtained from internet ).

© 2013 The Authors. Published by Elsevier Ltd.

Selection and/or peer-review under responsibility of the TerraGreen Academy

Keyword:Weibull distribution; Parameters of Wiebull; Wind energy; Wind turbine characteristique; operating hours

1. Introduction

Limited reserves of fossil fuels and their negative impacts on the environment lead institutions, organization and governments to find out more efficient technologies and new and renewable energy resources to produce energy in the natural environment. Recently, wind energy is the growing energy

* Corresponding M.DAHBI. Tel.: +0.213.655.489.93; fax: +213.49.81.52.44. E-mail address: dahbimaamar@yahoo.fr;univ .Bechar, Algeria, B.P 417.08000.

1876-6102 © 2013 The Authors. Published by Elsevier Ltd.

Selection and/or peer-review under responsibility of the TerraGreen Academy

doi:10.1016/j.egypro.2013.07.021

Nomenclature

2 <J Variance of wind velocity (m/s)

f ( V ) Probability of observing wind speed (%)

V Wind speed (m/s)

c Scale parameter (m/s)

k Shape parameter

V Mean wind speed (m/s)

Vc Cut-in wind speed (m/s)

VR Rated wind speed (m/s)

V F Cut-off wind speed (m/s)

Pr Rated electrical power (W)

tm Hours of months

P Average power (watt)

P Density of air (Kg/mA3)

A Area perpendicular to the wind ( mA2)

r Gamma function

Pw Power output of turbine (w)

p Operating hours.(Hours)

source of energy in the world and wind power is one of the most widely used alternative sources[1].The distribution of wind speeds is important for the desing of wind farms, power generators and agricultural applications such as the irrigation. It is not easy task to choose a site for a wind turbine because many factors have to be taken into account[1] . Compared to the other renewable energy resources, such as tidal or solar energy, wind energy has a more variable and diffuses energy flux. In order to maximise the benefit of this resource it is very important to be able to describe the variation of wind velocity at any given site under consideration for the development of WECTS.[2].The variation of wind velocity using the Weibull two parameter density function .

There are several methods to calculate the two Weibull parameters, c, and k such as the maximum likelihood method, the proposed modified maximum likelihood method and the commonly used graphical method [3-4]. Here the Weibull density function is used.

Algeria has a vast uninhabited land area where the south (desert) represents the part with considerable wind regime [5].

In Algeria, work on wind energy resource assessment dates back to 1976 when a wind atlas was developed by using wind speed data from 20 locations [6]. This atlas presented the monthly mean wind speed contour end frequency distribution for all the months during the year.

Dahbi and all [7] calculated the monthly and seasonally Weibull parameters for Bechar's site at 10m and found that the wind speed was well represented by the Weibull distribution function, based on wind speed data measured at 10 m height and collected along 12 years by the Algerian Meteorological Office.

This Paper present the monthly wind speed and wind power density to assess the wind power potential for the Sahara site of Algeria (South Western of Algeria). The Weibull density function has been used to estimate the two monthly Weibull parameters, c, and k.

The monthly power output of a wind turbine and the monthly operating hours for BWC XL.1 1KW wind turbine are calculated. Simulation is performed using Matlab software environment.

2. Simulation of wind power and wind turbine characteristics.

2.1. Weibull density function

f (v )=

The wind speed probability density function can be calculated as [8]: k

N f N k-1 - \ / V

■ exp

Where f (V ) ___,.__.._., ...... ^

parameter and k is the dimensionless Weibull shape parameter.

is the probability of observing wind speed v , c is the Weibull scale

Basically , the scale parameter, c indicates how 'windy' a wind location under consideration is, whereas the shape parameter, k , indicates how peaked the wind distrubution is (i.e. if the wind speeds tend to be very close to a certain value, the distribution will have a high k value and be very peaked)[9-10]. _ 2

Once the mean speed v ,, and the variance, CT , of data are known, the following approximation can be used to calculate the Weibull parameters , k and c :

/ N-1.086

Where the average wind speed V is - 1 n

V = —£ v

The variance ^ of wind velocity recordings

II (V, - V)2

And the gamma function of (x) (standard formula) calculated

-M x-1

r(x ) = J e~"ux ~ldu

The main limitation of the Weibull density function is that it does not accurately represent the probabilities of observing zero or very low wind speeds [10].Nevertheless, Persaud et al.[11] propose an alternative hybrid Weibull function as well as referring to the Dirac Dalta function allowing for probabilistic assessment of zero and low wind speeds.

2.2. The average power in the wind The average power in the wind can be expressed as [12]:

V 3 f (v )dv (4)

p = 2 pA¡ v 3f (v )dv

If is the Weibull density function, the average power becomes:

i + 3 k

1 +1 k

2.3. Power output characteristics of wind turbine

As different generators have different power output performance curves, so the model used to describe the performance is also different .In literature [13-14], the following equation is used to simulate the power output of wind turbine:

Pw (v) =

p * r R

(VC ^ V < V R )

(Vr < V < VF ) (V < v C & V > v F )

Where Pr is the rated electrical power ; Vc is the cut-in wind speed; Vr is the rated wind speed Vf is the cut-off wind speed of the wind turbine; k is the Weibull shape parameter.

2-4. The operating hours of the wind turbine

The cumulative distribution function can be used for estimating the time for which wind is within a certain velocity interval. The monthly operating hours in a year can be calculated like follow [15]:

Vc/ 1 I Vf/ 1

f ) = (e - e - tm (7)

p (vc < v < vF) = (e v c 7 - e v ,c} )* tm

Where Vc is the cut-in speed, Vf is the cut-of speed, tm is the monthly hours number (30 day chosen for each month) , k is the Weibull shape factor and c is the Weibull scale factor.

3.A case study for wind power potential analysis

The Adrar(South West Algerian) in new valley was selected as the site under consideration in this work since is the second largest province in Algeria, with an area of 427,368 km2. It is composed of 299ksour [16]. It is based around an oasis of the Sahara Desert .

A meteorological data collected during 8760hours by a wind observation station web site weather underground (The global whether data could be obtained from internet ) is used for analysis in this paper.

3.1. Yearly Weibull density function

The calculation results meet the Weibull distribution. From the recorded wind data , the shape parameter k is found to be 2.08, and the scale parameter c is 6.66m/s using Equ.1-3.The distribution is shown in figl.

Figl. Yearly wind probability density (k=2.43;c=6.66m/s)

3.1. Monthly Weibull parameters and wind power density

Once the monthly mean wind speed and the variance are known, the monthly probability density can be obtained, as shown in table 1. The results show that the parameters are distinctive for different months in a year, which means the monthly wind speed distribution differs over a whole year. It is clear that the mean wind speed increases during spring months and decrease during Autumns months. For comparison, tow typical monthly Weibull distribution with different shape and scale parameters are given in fig.2. It is clear from the figure that the value of c is low in September and most of the wind speeds are in the lower speed range, but in June the value is highest.

Table. 1: Monthly Weibull parameters and wind power density.

Months Mean wind speed (m/s) Shape parameter k Scale parameter C (m/s) Avrage power P (W/m2)

January 5.22 2.58 5.88 142.89

February 6.21 2.83 6.98 226.85

March 6.28 2.27 7.10 274.84

April 5.97 1.86 6.68 278

May 6.65 2.15 7.51 340.97

June 7.30 2.29 8.25 428.040

July 5.14 1.87 5.79 181.28

August 5.86 2.67 6.59 197.58

September 4.45 1.72 4.99 129.22

October 6.28 2.87 7.05 232.38

November 4.92 2.01 5.56 147.48

December 5.82 3.33 6.48 192.99

Figure 2 : The Weibull distribution in June and September

3.2. Monthly wind power density

Using Eq.(5), the average annual wind power density is found as 231w/mA2 , so that the power potential is 1774.08 kWh/m2 per year . With the above equation and the monthly Weibull parameters, the monthly wind power density can also be calculated. The results are given in fig.3 and tabl.1

It is clear from the results that the mean wind speed and average wind power density are distinctive for different months. In February to June, the wind power is high, but low in the autumn months. The tow

curves have similar changing trend, but the rate of change is not the same. This is because, for the power density, it is determined not only by the mean speed bat also by the Weibull shape parameter k.

Fig.3: Monthly wind power density and .mean wind speed

3.3. Monthly wind power density

To analysis the effect of the wind turbine characteristics on the wind power generation, a wind turbine of BWC XL.1 1KW generator with a synchronous permanent magnets machine is chosen in this study based on the calculated monthly mean wind speed (is high to 2.5 and low to 13m/s). The cut-in speed is 2.5m/s , the cut-off speed is 13m/s and the rated electrical power of the turbine is 1kW(at rated wind speed 11m/s).Using the eq.6 and Eq.7 ,the average monthly power generated by the wind turbine and the operating hours of the wind turbine in a year can be calculated, as shown in fig .4. it is clear from the figure that the operating hours changes in different months. The highest 690h occurs in December, while the lowest is 527h in September. The total operating hours of the Adrar site are 7621h in a year .the trend is not in accordance with the trend of monthly wind power density, because the operating hours is determined not only the wind data distribution but also by the performance of the wind turbine. For different months, the monthly power generated is calculated and the results are given in fig4.The results shows that for each month the power generated is different , varying from 113.54 W to 370.58W .The highest power output is generated in June , while the lowest in September. The wind turbine can generate electrical power in the site of Adrar amounting to 78410W (1559.400kWh) per year. The trend is similar to that for power density, but with some difference. This is because the power generated by a wind turbine is determined not only by the wind speed's Weibull function (parameters c and k),but also by the characteristics of wind turbines, while power density is determined only by the weibull distribution (c and k).

Fig.4: Monthly operating percentage and monthly power output of the wind turbine

Conclusion

The proposed simulation model is found applicable for assessing the potential of wind power generation at a location. From the case study, it is found that it is imperative to account for the monthly difference of wind speed using Weibull estimation as the wind energy potential can vary significantly. According to result's of simulation, it is shown that the parameters of Weibull influence directly the wind power available like over the duration of production of electrical energy . In June, the wind power available in 428 W/m2with one duration of availability of 21 h per day. In September, the wind power available is 128 W/m2with one duration of availability of 17h per day.

In the case of Adrar, it is fortunate that the windy season coincides with the high temperature seasons (45°-55°) which causes an increased demand for electricity could be covered by exploitation of wind energy. Therefore, accurate wind predictability and its correlation with electricity load demand may allow for a high penetration of wind energy and could make it an economically attractive supplement to diesel-fuelled power station.

It is recommended to undertake further studies to explore other locations in the Sahara of Algeria and to develop this renewable energy resource.

Reference

[1] Mathew, S. ; Pandey,K. P., and Kumar, A.: Analysis of Wind Regimes for Energy Estimation, Renewable Energy, 2002 ;25(3): 381-399

[2] Ulgen K., Hepbasli A..: Determination of Weibull parameters for Wind Energy analysis fo Ismir, Turkey. Int J Energy Res.2002; 26(6):495-506.

[3] Lun I.Y.F; Lam JC. .: A Study of Weibull Paramétré using Long -term Wind Observations. Renew Energy 2000; 20(2): 145-

[4] Seguro JV.; Lambert TW. .: Modern Estimation of the Parameters of the Weibull wind speed Distribution for Wind Power

Energy Analysis .J Wind Energy Ind Aerodyn 2000; 85(1):75-84.

[5] Diaf,S.; Belhamel, M.; Haddadi, M.; Louche, A.: Assesment of Wind Energy Ressource in Southern Algeria, Rev.Energ. Ren.

2000; 20: 145-153.

[6[ Hammouche,R. : Wind Atlas of Algeria, Meteorological National Office, University Publications Office. Algiers, Algeria, 1991.

[7 ] Dahbi M. ; Benatiallah A. ; Bbenslimane T. : Analyse du Potentiel Eolien dans Le Site de Bechar : Caractérisation des Paramètres de Weibull , The First International Workshop on Renewable Energies and their Applications (1 WIERA) ,Laghouat, Algeria, May 10-12, 2008: 231-236. [8] Gary L Johnson : Wind Energy Systems . Manhatan KS.; November 20 2001: 30-38.

[9] Justus, C. G. : Wind and Wind System Performance. Franklin Institute Press; Philaelphia;1978.

[10] Weisser D : A Wind Energyr Analysis of Grenada: An Estimation Using the 'Weibull' Density Function, Renew Energy, , 2003; 28.: 1803-1812

[11 ] Persaud S.; Flynn D., Fox B.. :Potential for Wind Generation on the Guyana Coastlands, Renew Energy; 1999; 18(2).: 175— 189.

[12] Lin,L.; Hongxing,Y.; John,B.:Investigation on Wind Power Potentiale on Hong Kong Island-an Analysis of Wind Power and Wind Turbine Characteristics, Renew Energy 2002; 27,:1-12..

[13] Ghali FMA., Abd el Aziz MM..; Syam FA.:Simulation and analysis of Hybrid system Using Probabilistic techniques. In ; Power Conversion Conference; Nagaoka ,1997,2;831-835 .

[14 ] Gavanidou ES.; Bakirtzis AG. .:Design of a Stand Alone System with Renewable Energy Sources Using trade of methods. IEEE , Trans. Energy Conversion, 1999;7(1),;;42-48.

[15] Hennessey JP. : Somme Aspects of Wind Power Statistics .J Applied Meteorology; 1977, 16:119-128.

[16] Laaboudi A. ; Slama A ; Hamouda M. : Impact des Changements Climatiques sur les Ressources en eau dans la Région

d'Adrar ; International Conference on Energy and Sustainable Development 29-30 November 2011 .