Wind characteristics of

* ** *** Y. Himri , B. Draoui and S. Himri

* Electricity & Gas National Enterprise (Sonelgaz), Béchar. Algeria 05 rue Mokadem Ahmed, Béchar (08000) Algeria, [email protected] ** University of Béchar, Algeria, [email protected] *** University of Béchar, Algeria, [email protected]

ABSTRACT

The paper presents wind speed data analysis over a period In Algeria work on wind resource assessment dates of almost 10 years between 1977 and 1988 from three back to 1990 when a wind atlas was published by stations, namely Adrar, and Tindouf. The long- Hammouche [13] giving the results of the statistical study term mean wind speed at these sites was 5.9, 5.1 and 4.3 m/s of 37 locations using the WASP. Kasbadji Merzouk at 10meters above ground level (AGL), respectively and performed mean wind data from 64 stations for wind registered every three hour interval at all the stations. Higher energy potential of Algeria at a height of 10 m. It is found wind speeds were observed in the day time between 09:00 that the windy regions are located at the south west of and 18:00 hours and relatively smaller during rest of the Algeria, [14]. period. The data show also that the long-term seasonal wind Ettoumi et al used first-order Markov chain and speeds were found to be relatively higher during March to Weibull distribution methods for statistical bivariate September compared to other months so which helps in modelling of wind using the data wind speed and wind planning the operation of existing power plants in direction measurements collected every 3 h at the coordination with the wind power, since that the energy meteorological station of Essenia, also a detailed study could be harnessed for almost 64 % the time using wind has been made on the statistical features of the wind at machines with cut-in-speed of 3 m/s or more. (Algeria) [15]. Himri et al utilized wind speed data over a period of Keywords: Algeria, frequency distribution, wind characteristic. almost 10 years from three stations, namely Adrar, Timimoun and Tindouf to assess the wind power potential 1 INTRODUCTION at these sites, they found Wind farms of 30MW installed capacity at Adrar, Timimoun and Tindouf, if developed, Currently, wind energy is generally developed in the could produce 98,832, 78,138 and 56,040MWh of industrialized countries for a number of reasons including electricity annually [16]. growing environmental concerns and relatively high price of This paper presents the wind data analysis at three electricity generated. The total world wind power capacity locations; namely Adrar, Timimoun and Tindouf in Algeria. reached up72 628 MW in 2006. Wind power is present today The Weibull wind distribution was calculated using the in the energy mix of more than 60 countries, not only in Excel application program to generate graphs. practically all of the developed countries but also in more and more of the developing ones. India already has a 6 053 MW 2 WIND DATA AND SITE DESCRIPTION installed capacity, which represents the 4th place position behind Germany, Spain and the USA. China enters the top The wind speed data were collected over a period of 8 ten (in 8th place) with 1 699 MW installed capacity [1]. to 10 years at three sites used in this study. The latitude, With growing global awareness of the usage of clean sources of longitude, altitude, measurement duration and years of energy, wind energy in particular, a lot of work is being carried measurements are summarized in Table 1. out in different parts of the world, as can be seen from [2-12].

Location Latitude (deg) Longitude (deg) Altitude (m) Duration (years) Measurement Years Adrar 27° 49’ N 00° 17’ E 263 11 01/01/1977 to 07/12/1988 Timimoun 29° 15’ N 00° 17’ E 312 11 01/01/1977 to 31/12/1988 Tindouf 27° 40’ N 08° 06’ W 401 08 01/01/1976 to 01/12/1984

Table 1: Geographical coordinates of data collection stations used in the study.

Clean Technology 2008, www.ct-si.org, ISBN 978-1-4200-8502-0 133

The wind speed measurements were made 10 meters Tindouf Timimoun Adrar above ground level and recorded every three hour interval 8 (viz. 0, 3, 6, 9, 12, 15, 18 and 21 hours) at all the stations. 7 The wind data measurement stations lie in the South west 6

region of Algeria. The geographical locations of these 5

stations are also shown in Figure 1. 4

3

2 Mean wind speed, (m/s)

1

0 0 3 6 9 12 15 18 21 Hour of the Day

Figure 3: Diurnal variation of long-term mean wind speed.

In general, higher winds were observed between 09:00 and 18:00 hours while lower during rest of the period of the day. This indicates that higher electricity could be produced during 09:00 to 18:00 hours which also coincide with higher electricity demand time. It is also evident from this figure that wind speed was highest at Ardar while lowest at Tindouf during entire day. The seasonal variation of wind speed provides information about the availability of wind speed during different months of the year which helps in planning the operation of existing power plants in coordination with the wind power plant. Figure 1: Map showing the locations of the wind As shown in figures 4 - 6 the frequency distribution of measurement stations. wind data indicates that wind remained 4 m/s and below it for about 25% ,33% and 35% of the time during the entire Figure 2 provides the variation of monthly average wind year at Adrar, Timimoun and Tindouf respectively and speed during entire data collection period at three stations above it for rest of the period. under consideration in this study. Again, highest wind This means, that the energy could be harnessed for 65% speeds were found at Adrar whiles the lowest at Tindouf to 75% of time using wind machines with cut-in-speed of 3 during the year. The monthly mean wind speed was in m/s or more at these locations. between the two stations, as seen from figure 2. 14

Adrar Timimoun Tindouf 12

Best-fit Weibull 10 7 (k=2.15, c=7.2) ) 6 8 5

4 6

3 (% Frequency 4 2

1 2 Mean wind speed, (m/s) speed, wind Mean 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0 Month 1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 Wind speed (m/s)

Figure 2: Seasonal variation of long-term mean wind speed. Figure 4: Frequency distribution hourly mean wind speed The diurnal change in long-term mean wind speed at in Adrar. Adrar, Timimoun and Tindouf stations is shown in Figure 3. The maximum wind speed was found to occur at 09:00 hours at all the locations while the minimum at 21:00 hours.

134 Clean Technology 2008, www.ct-si.org, ISBN 978-1-4200-8502-0

REFERENCES 14 [1] GWEC Global Wind Energy Council Press Release,

12 “Global wind energy markets continue to boom," 2006. 10 Best-fit Weibull [2] S.Rehman, and T.O.Halawani, “Statistical (k=1.91, c=6.5)

) 8 Characteristics of Wind in Saudi Arabia,”

6 Renewable Energy, 4(8), pp. 949-956, 1994. [3] S.Rehman, T.O. Halawani, and T. Husain, “ Weibull 4 Parameters for Wind Speed Distribution in Saudi Frequency (% Frequency

2 Arabia,” Solar Energy, 53(6), pp. 473-479, 1994. [4] H.S. Bagiorgas, M.N. Assimakopoulos, Theoharopoulos, 0 1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 D. Matthopoulos and G.K. Mihalakakou, “Electricity

Wind speed (m/s) generation using wind energy conversion systems in the area of Western Greece, “ Energy Conversion and Figure 5: Frequency distribution hourly mean wind speed Management 48, pp. 1640–1655, 2007. in Timimoun. [5] J. Yingni, Y. Xiuling, F. Jianmei and C. Xiaojun,

“Wind Power Density Statistics Using the Weibull 16 Model for Inner Mongolia,” Wind Engineering 14 30(2), pp. 161–168, 2006.

12 Best-fit Weibull [6] K. Ulgen, A. Genc, A. Hepbasli, et al., “Assessment (k=1.98, c=6.1) of wind characteristics for energy generation,” 10 Energy Sources, 26 (13), pp. 1227-1237, 2004. 8 [7] R.K. Panda, T.K. Sarkar, A.K. Bhattacharya, “Stochastic study of wind energy potential of Frequency (%) 6 India,” Energy, 1990. 4 [8] M. Jamil, S. Parsa, M. Majidi, “Wind power

2 statistics and an evaluation of wind energy density,” Renewable Energy, 1995. 0 12345678910111213141516171819202122232425 [9] T.V. Ramachandra, D.K. Subramanian, N.V. Joshi, Wind speed (m/s) “Wind energy potential assessment in Uttara Kannada district of Karnataka, India,” Renewable Figure 6 : Frequency distribution hourly mean wind speed Energy, 1997. in Tindouf. [10] S.H.Alawaji, “Wind energy resource assessment in Saudi Arabia,” network design and description. Renewable Energy, 1996. 3 CONCLUSION [11] A.E. Feijoo, J.Cidras, J.L.G Dornelas, “Wind speed simulation in wind farms for security assessment of The present study performed using the long-term annual electrical power systems,” IEEE Trans Energy mean wind speed data at three locations, viz. Adrar, Convers, 1999. Timimoun and Tindouf in Algeria. The study found the [12] Z. Sen, A.D. Sahin, “Regional assessment of wind following salient features as the outcome of the analysis: power in Western Turkey by the cumulative semivarigram method,” Renewable Energy 1997. • The long-term mean wind speed at Adrar, [13] R. Hammouche," Atlas Vent de l’Algérie," Timimoun and Tindouf was 5.9, 5.1 and 4.3 m/s at publication interne de l’ONM, (Office National de 10 meters above ground level (AGL). Météorologie), Alger, 1990. • The higher wind speeds were observed in the day [14] N. Kasbadji Merzouk, “Wind energy potential of time between 09:00 and 18:00 hours and relatively Algeria," Renewable Energy, pp 553-562, 2000. smaller during rest of the period. [15] F. Youcef Ettoumi, H. Sauvageot, A.E.H. Adane, • The long-term seasonal wind speeds were found to “Statistical bivariate modeling of wind using first- be relatively higher during March to September order Markov chain and Weibull distribution,” compared to other months. Renewable Energy, 2003. • The electricity could be produced for 75 %, 67% [16] Y. Himri, S. Rehman, B. Draoui and S. Himri, and 65% of the time at Adrar, Timimoun and “Wind power potential assessment for three Tindouf respectively. locations in Algeria,” Renewable and Sustainable Energy Reviews, 2007.

Clean Technology 2008, www.ct-si.org, ISBN 978-1-4200-8502-0 135