Available online at www.sciencedirect.com

SciVerse ScienceDirect

APCBEE Procedía 5 (2013) 451 - 467

ICESD 2013: 19-20 January 2013, Dubai, UAE

Estimation of Electricity Generation in Libya Using Processing Technology of Wind Available Data: The Case study in Derna

F.Ahwide , A.Spenab and A. El-Kafrawy

a Omar Al-Mukhtar University -Darnah-Libya-Faculty of Engineering, Mechanical Department, b Universitä di Roma Tor Vergata - Facoltä di Ingegneria, Dipartimento dilngegneria dell'Impresa, cAl-Baha University-Faculty of Engineering-Mechanical Engineering Dept., KSA Port-Said University-Faculty of Engineering-Production Engineering and Machine Design Dept.

Abstract

This paper is concerned with the processing technology of wind available data as a means to estimate the electricity generation in Derna site - Libya, which are located on the coast of Mediterranean Sea. The work presented long term wind data analysis in terms of annual, seasonal and diurnal variations at these sites. The wind speed and wind direction each 3 hours data for a period of 10 years between 2000 and 2009 were used in the analysis. Weibull parameters at different heights has been studied and used to describe the distribution and behavior of wind speed and their frequencies. Wind turbine energy production at different heights has been calculated using a Weibull distribution (Weibull Statistics Techniques). The total energy output per year from the wind machines of different sizes and at different hub heights were compared. This research work revealed that the lower rated speed for the wind energy conversion system (WECS) of the same height, the higher the capacity factor values, in most cases. This work used also a new analytical method of (Spena & Ahwide's method) for the estimation of the available wind energy potential. The results of both previous mentioned method's were compared. Moreover, the expected cost in $ Dollar/kWh for both wind machines have been calculated, which have different characteristics but the same power level of 1650 kW. It was hoped that the data analysis can help to identify good sites in Libya for new wind turbine installations. This evaluation was helpful to trigger the use of large wind turbines at the selected sites along the coasts of Mediterranean.

© 2013 The Authors. Published by Elsevier B.V.

Selection and peer review under responsibility of Asia-Pacific Chemical, Biological & Environmental Engineering Society Keywords: Wind Data, Electricity Generation in Libya, Wind Turbines, Weibull Statistics Techniques

* Corresponding author. Tel.: 00966556404890. E-mail address: dr_eng_aly@hotmail.com

2212-6708 © 2013 The Authors. Published by Elsevier B.V.

Selection and peer review under responsibility of Asia-Pacific Chemical, Biological & Environmental Engineering Society doi: 10.1016/j.apcbee.2013.05.078

1. Introduction

In this era of technological advances and materialistic life style, energy has become an essential entity for inhabitants of the planet. In this modern world there are more than a billion people who have no access to electricity in various parts of the globe. Most of these unfortunate people are living in developing countries. Beyond doubt, one must use energy more efficiently. In the coming times the developing world will need more energy to address its essential needs. The challenge that all of us are facing is how to meet this growing demand of energy while at the same time addressing the equally urgent threat of climate change. To address the pollution problem, green sources of energy like solar, hydropower, wind, tidal, biogas, wave energy, etc. are being encouraged. Of these green sources, the usage of wind as a source of energy is increasing in different parts of the globe due to rapid technology advancement. Wind energy utilization is also becoming competitive compared to traditional sources of energy. This paper presented the detailed wind data analysis and wind availability at Derna city which Located on coast of Mediterranean Sea in Libya. Fig. 1 shows the location of Libya in North Africa.

2. Wind data analysis and resource estimation

2.1. Mean wind speed

2.2. Estimation of wind and data analysis

The wind data were obtained from the representative meteorological station. Long term time series, for 10 years, of 3-hourly measured wind data were used. Fig.(2) shows the monthly variation of the average wind speed for long period of time at 10 m height, the minimum value of average wind speed was in May and the maximum in Dec. The mean wind speed for 10 years was 5.67 m/s [1,2].

Fig. 1. Location of Libya in North Africa

Fig. 2. Average monthly wind speed every 3 hours, Dernah-Libya

2.3. Wind speed variation with height

In wind power studies, two mathematical models have been used to represent the vertical wind speed over homogeneous areas are as follows.

2.3.1. Power Law profile

The wind speed as a function of height according to the power law profile has the form

V(z) = V(Zr)(—)a (2)

Where V(z) is the wind speed at height z, V(z) is the reference wind speed at height zr ,and a is the power law exponent.

2.3.2. Logarithmic profile

The wind speed as a function of height according to the logarithmic profile is

V(z) = V(zr)*[ln(—)/ln(-^)] (3)

Where V(zr) is the wind speed at height zr above the ground level, and z0 is the roughness length.(Hiester and Pennell, 1981) [3 , 4]. Fig.3, shows mean wind speeds at certain heights 20, 40, 60, 70, 100 and 125m. We observed significant changes in wind speed with different heights.

3. Statistical analysis of wind data

3.1. Estimation of Weibull parameters

Graphical method was used to estimate the Weibull parameters from the obtained data :

1-F(V) = exp_(V/C)k (4)

ln[—ln(1 — F(V))] = kln V-k ln C (5)

3.2. Weibull distribution

In Weibull distribution, the variations in wind velocity are characterized by two functions: (1) The probability density function. (2) The cumulative distribution function. It is given by using two equations, [5 , 6].

F(V) = C ("C^"1 exp_(V/C)k (6)

F(V) = 1-exp[-(V/C)k ] (7)

Where F(v) is the Weibull cumulative distribution function, k is the Weibull shape parameter and c is the scale parameter (m/s). Fig. 4 shows the technique used to determine these parameters for Darnah site, at different heights, years from 2000 to 2009.

Fig. 3. Variation of annual mean wind speeds every 3 hours with different height (Darnah, years from 2000 to 2009) Table 1. Numerical values of height h, scale parameter c, and weibull shape parameter k

Height (m) c (m/s) k

10 5,9 1,55

20 6.6 1,56

30 6,8 1,56

40 6,9 1,55

50 7 1,54

60 7,1 1,53

70 7,5 1,56

80 7,6 1,56

90 7,6 1,56

100 7,7 1,56

125 7,8 1,55

Fig. 4. Graphical determination of weibull parameters at different height, Darnah City

Fig. 5. Probability distribution of average wind speed every 3 hours at different heights

Fig. 6. Cumulative weibull distribution at different heights

Fig. 7. Energy output at various heights

Weibull distribution and probability distribution of the year for Derna station, as shown in Fig. (5) and Fig.(6).

3.3. Estimate of energy production using a Weibull distribution

In this study Weibull distribution statistics has been used that was by compensation in the equation(8) to calculate the average wind machine power and energy output, the wind duration availability in terms of number of hours the wind remained in a particular bin is calculated by constructing the wind duration distribution[5].

=Z exp

V + Vj 2

4. Rated wind speed and capacity factor of WECS

The following analysis is to help designers and users to choose the most suitable wind turbines. The annual energy products of some different commercial wind turbines (each two of them have the same hub height and the same rated power but are different rated wind speed (Vr)). The rated powers of these turbines were 600, 900, 1000, 1500 and 1650kW. Table (2) shows their annual energy productions, Eout, and capacity factors, where capacity factor is the ratio between the actual yearly energy output, Eout, and the rated yearly energy. From the results in the table 2, which are presented in Fig. 7, we concluded that the capacity factor is greater for wind turbines with lower rated wind speeds.

5. Cost Analysis

The estimation of the cost per kWh of energy produced by the wind turbine "Vestas V82", which has a capacity of 1650 kW, to operate at Derna station has been done under the following assumptions:

* Investment (I ) includes the turbine price plus its 20% for the civil work and other connections.

* Operation maintenance and repair cost (Comr) were considered to be 25% of the annual cost of the turbine (machine price/lifetime).

* The interest rate (r) and inflation rate (i) were taken to be 15% and 12%, respectively.

* Scrap value S was taken to be 10% of the turbine price and civil work.

* The lifetime of the machine (t) was assumed to be 20 years.

* the present value of costs (PVC) is [8 , 19]:

PVC = 1 + Co

"1 + i" * "1 + i" t" -S "1 + i"

_r-i_ _1 + r_ _1 + r_

Technical data of the chosen wind machine is given in Fig.9. By substituting these values in above Eq. with above assumptions we obtain the PVC for each location. The cost of electricity per kWh at each location is obtained by dividing the PVC by the total energy production of the wind turbine over its life, time (20 years).

The price of this turbine Vestas V82-1.65MW is taken to be $3,750,000. The cost of civil work (20% of the price) = $750,000. Therefore, investment I = $4,500,000, Comr = $(3,750,000/20)*0.25 = $46,875 S = $(4,500,000*0.1) = $450,000, where r = 0.15 i = 0.12. Using all these values in above Eq. [7], we get PVC = 4,953,507.5.

Also from Table 4, the annual output of turbine Vestas V82-1.65MW at Darnah is 6,645,740 kWh. So the total output over 20 years = (20*6,645,740)kWh. Hence, the specific cost per kWh = [(4,953,507.5)/ (20*6,645,740)] = 0,38$ dollar.

While, when compared to "mtORRES (TWT 1.65/82)" we conclude that the specific cost per kWh = [(5,267,106)/ (20*6,523,040)] = 0,41$ dollar.

Hence, we note that the specific cost per kWh for" Vestas V82" Less than a "mtORRES (TWT 1.65/82) ".

Table 2. Technical data of wind machines used in the analysis [10, 17]

cut in cut out Rated Rated power Turbine Model speed speed speed (kW)

(m/s) (m/s) (m/s)

Hub height Rotor Energy Cf (m) diameter output (%) (m) (kWh/y)

Enercon E40-600kW 2.5 28-34 12.5 600 40 40 1,204,042 23

AN BONUS 41-600kW 4 25 14-15 600 40 41 1,087,883 21

AWE52-900 kW 2 25 14 900 75 52 2,606,238 33

NM 52-900 kW 3.5 25 15 900 75 52 2,522,698 32

FL 54-1000 kW 2.5 25 12.5 1000 70 54 2,873,724 33

aaER A-1000/S 3 20 14 1000 70 54 2,657,101 30

HW77/1500 kW 3 25 12 1500 70 77 6,078,455 46

GE 77-1.5MW 3.5 25 14 1500 70 77 5,878,455 45

V82-1.65MW 3.5 20 12 1650 80 82 7,076,782 49

TWT 82-1.65 MW 3 25 14 1650 80 82 6,943,509 48

Table 3. Energy produced using a Webull distribution of different sizes and different hub height, Darnah-Libya

Wind Speed m/s hours year h=80m TWT 1.65/82 (kWh) hours year h=70m V-82 1650kW (kWh) TWT 1.65/82 (kWh) HW-77 1500kW (kWh) GE-77 1500kW (kWh) TWT 1.65/70 (kwh)

1 783 0 783 0 0 0 0 0

2 175 0 175 0 0 0 0 0

3 195 0 195 0 0 0 0 5

4 1198 3340 1198 1498 3370 1532 2043 1392

5 908 15421 908 14624 15538 8727 9697 7237

6 67 4374 67 4655 4399 2877 2877 2214

7 1330 246387 1330 261589 247255 234622 219958 132118

8 776 326548 992 442199 417697 445623 389920 237800

9 460 369743 245 203917 196097 201025 172307 119790

10 1052 1378959 1052 1436057 1372357 1333852 1280498 908256

11 116 215499 635 1234507 1175408 1136430 1097243 840222

12 548 1290415 29 70335 66984 61538 59487 51238

13 523 1413003 523 1419786 1391390 1232422 1224206 1114673

14 201 586348 233 667850 664207 579716 579716 540726

15 61 181830 29 86206 86154 74830 74830 69919

16 130 386083 148 429276 429276 372626 372626 343096

17 33 95554 15 40824 40824 35437 35437 31905

18 48 129102 48 124346 124346 107936 107936 94469

19 77 188056 77 182338 182338 158275 158275 133963

20 13 28969 12 25734 25734 22338 22338 18189

21 11 20814 11 0 19735 17038 17038 13278

22 23 38373 26 0 39428 34224 34224 25395

23 4 5366 2 0 2322 2015 2015 1417

24 5 6119 7 0 6970 6050 6050 4008

25 14 13207 13 0 11212 9733 9733 6046

8760 6943509 6645740 6523040 6078868 5878455 4697353

6. Assessment of Energy Output By Other Method (Spena & Ahwide method's)

6.1. A case study, (Dernah, for years from 2000 to 2009).

Derna city was selected because the wind speed is acceptable and it was proposed a project of construction of wind resource analysis tender Derna farm. The mtORRES (TWT 1.65/82) is the industrial wind turbine used for this project. This project was intended to study the wind resource affecting Tender Dernah wind farm in Libya. The project consists of three possible scenarios: 60MW (37WTG), 70MW (43WTG) and 120MW. Tender Dernah Wind Farm will be located one kilometre away from the coast, in the surrounding of Dernah. Libya see Figs. 8-9 bellow.

Fig. 8. Dernah Wind Farm

Fig. 9 Wind farm positions

6.2. Wind power density

The wind power density was calculated from equation

1 3 P = -pAV3 2

This has rendered possible to design the diagrams of average wind speed duration (V) (see Fig.10) and of

26 24 22 -

10 20 JO 40 50 <1 % 1 1 1 0 70 80 90 1 JO

Fig. 10. Duration of average wind speed

Fig. 11. Specific power durations Derna- Libya 6.2 load factor and Energy yield

The annual energy and annual capacity factor were calculated at 70m height based on specification of wind turbine known as mtORRES (TWT 1.65/82) wind turbine which has a power curve as shown in Figure 12, and its main characteristics are shown in Fig 17. Where the air density p=1.225 kg/m3. The power curve of a industrial wind turbine shows the typical trend of the wind speed. The power was equal to zero up to a speed of about V=3.0 m/s and reached 100% at speed V=15.0 m/s and the cut-out speed is 25.0 m/s.

Fig. 12. Power curve of wind turbine

Fig. 13. Curve of Load Factor of mtORRES (TWT 1.65/82)

From the curve of Fig. 12 the load factor (L.F) was calculated (see Fig.13), which is defined as the ratio between the instantaneous power and the max. load of wind turbine, in relation the specific power P' .corresponding to each wind speed (V) according to equation of wind power density. In Fig. 14 the curve of (L.F) was also superimposed on the distribution frequency of wind velocities of Derna-Libya at 70 m height. Note, the Proportionality between the power output and the relative frequency distribution.

c <-.11

n J n n

IL Ii M J n n n

]] 13 15 17 19 21 23 25

V(1D 5)

Fig. 14. Curves of Load Factor and distribution frequency of wind velocities of Derna-Libya

Fig. 15. Specific power (P') and the load factor (L.F.)

Fig. 15 shows the curves of the specific power (P') and the load factor (L.F.) as a function of the coefficient u. Multiplying each value of LF(u) for the duration (u) in function the correspondent specific power P'(u), then, one can obtain a function of integral energy yield (RI) of the wind turbine generator, Where

RI= L.F * u(P') (see Fig. 16) [20].

6.3. Results and discussion

6.3.1. Annual energy output

Fig. 16 shows the optimal value of the integral energy yield for Darnah, which can be obtained through the corresponding value of specific power P' in around of 820W/m2/y. It gave us the average annual energy of approximately 18 % from the limit of a wind turbine generator. Therefore, the annual coefficient of the conventional use of time for Darnah was about 1576 hours (u= 0.18*8760) of the equivalent fully loaded operation. From above, it can be concluded that the energy yield was equal to the specific power (P') multiplying per coefficient of time (u). (Energy yield = 1293 kWh/m2/y), multiplying per swept area(A), the annual energy output was obtained as Eout = 6,936,945 kWh.

S|m ltk power & Moment <>t [1* 111

3(1(10 2750

—fr—p- —*-LF I.F" 1

S 1750 ■ £ 15(10 ■

(I 1« 2(1 3(1 4(1 50 Ml 70 SO 'Ml 10(1 u(%), LF(%)

Fig. 16. Specific Power and Moment of (L.F*u)

Power curves of (TWT 1.65/82>&(V82-1.65MW)

0 2 4 6 8 10 (2 14 16 IS 2D 22 24 26 Wind spt'l-ll [Ill s]

Fig. 17. Power curves of (TWT 1.65/82)&(V82)

6.3.2. Rated wind speed and capacity factor of WESC

> The annual energy output from the wind machine of mtORRES(TWT 1.65/82) at different sizes and at different hub heights has been studied (see tables 2,3).

> Energy output and capacity factor of different commercial wind turbines: (each two of them have the same hub height, rotor diameter and rated power but they are different rated wind speed) were estimated (see table 4 & Fig 18).

> The comparison between the both methods.

After studying the above items, it can be concluded that :

From (Tables 2, 3), the energy output of wind machines increases with the increasing of hub height, rotor diameter and rated power. This showed that hub height and rotor diameter play a considerable role in energy generation from the wind machines. Similar types of behavior are noticed from the wind machines of other

From the results (Table 4.7), the use of a wind turbine which has a rated power greater than 1000 kW at Darnah station was recommended. Then based on results (Fig.18), it was concluded that the use of a wind turbine with lower rated speed will produce speed. And from Table .4, the capacity factor is greater for wind turbine with lower rated wind speeds.

In this section , the choice of the best suitable wind turbine for each stations was discussed. Whereas a technical and economic assessment of electricity generation from turbine machine "mtORRES (TWT 1.65/82)" has a capacity of 1650 kW considered in ten different sites along the coast of Mediterranean Sea and Sahara in Libya (Darnah station was one of them). The results of two methods were compared and showed no difference between them (see Fig 19).

Fig. 18. Specific Power and Moment of (L.F*u)

Fig. 19. Comparison between the both method

7. Recommendations and conclusions

As for the wind energy, so the purpose of this part is to present a new analytical method for the calculation of the wind available power for different sites in Libya, which are located on the coast of Mediterranean Sea, central zone and Sahara and estimate the electrical power generated by large wind turbines and compared with Weibull statistics and moreover and evaluated the expected cost in € cent/kWh for the power level of 1650 kW and comparison with other machine turbine showed the same characteristics. After studying this part of work, it can be concluded that the energy output of wind machines increases with the increasing of hub height, rotor diameter and rated power. This showed that hub height and rotor diameter play a considerable role in energy generation from the wind machines. Similar types of behavior are noticed from the wind machines of other sizes. The use of a wind turbine which has a rated power greater than 1000kW at Darnah station was recommended. Then based on results, the use of a wind turbine with lower rated speed will produce more energy over a year than a wind turbine with higher rated speed. The capacity factor is greater for wind turbine with lower rated wind speeds. The choice of the best suitable wind turbine for each station was discussed. Whereas a technical and economic assessment of electricity generation from wind turbine machine "mtORRES (TWT 1.65/82)" has capacity of 1650kW considered in ten different sites along the coast of Mediterranean Sea and Sahara in Libya (Darnah station was one of them). The energy yield and capacity factor for Dernah on the coast of Mediterranean Sea was high and acceptable. The results of two methods were compared, hence we noted that there weren't difference between them. It is hoped that the data analysis will help to identify good sites in Libya for new wind turbine installations. This evaluation is hoped to trigger the use of large wind turbines at the selected sites along the coasts of Mediterranean

References

[1] Libyan National Meteorological center, Climatical Department, Tripoli, 2009.

[2] Libyan National Meteorological center, Climatical Department, Darnah, 2010.

[3] Estimation and assessment of wind energy in some areas in Libya, GCREEDER, Amman-Jordan, 31-3-2009.

[4] Mohamed, A. Elmabrouk, Assessment of the wind Energy Potential on the Coast of Tripoli.

[5] John Wiley & Sons, Wind Energy Explained, Theory, Design and Application, Copyright [2002 ISBN 0471499722 ].

[6] John, W. & Nicholas, J., Wind Energy Technology, Unesco, Paris, France translated in Arabic, Dr. Widad, A.

[7] Rodolfo Pallabazzer, SESTEMI EOLICI.

[8] A.S. Ahmed Shata, R. Hanitsch, Renewable Energy 33 (2008) 141-148, Electricity generation and wind potential assessment at Hurghada, Egypt.

[9] S. Rehman / Energy Conversion and Management 45 (2004) 2019-2032, Wind energy resources assessment for Yanbo, Saudi Arabia.

[10] (10 x Used Enercon Wind Turbines E40 - 600kW For Sale MyWindPowerSystem.com).

[11] RE power Systems AG 5M.mht.

[12] Mtorres TWT 165-82 in pdf.

[13] Enercon 82-2000kW Produktuebersicht_eng.

[14] GE Energy - 1.5 MW Series Wind Turbine.

[15] 1.5mW SAIP S77 Wind Turbine System - On Grid Products.

[16] Americas Wind Energy Inc. Home.

[17] Dewind_D80-2000kW_A4_small.

[18] General Information Authority, Libya, www.gia.gov.ly.

[19] Alnaser WE. Assessment of the possibility of using three types of wind turbine in Bahrain. Renew Energy 1993;3(2-3):179-8.

[20] F. Ahwide, A. Spena,IV Congresso Nazionale AIGE Roma,26-27 Maggio 2010, Estimation and assessment of wind energy potential in some areas in Libya.

Table 4. Annual energy output and capacity factor of 16 different commercial wind turbines(each two of them have the same hub height, rotor diameter and rated power but they are different in their rated wind speed:

rbine Model Hub height (m) Rated power (kW) Vcut in (m/s) Rated Speed (m/s) Vcut out (m/s) Energy Cf (kWh/y) (%)

Enercon E40 40 600 2.5 12.5 28-34 1,219,717 23

AN BONUS 41 40 600 4.5 14-15 25 1,167,883 22

AWE52-900 kW 75 900 2 14 25 2,883,967 37

NM 52-900 kW 75 900 3.5 15 25 2,669,995 34

FL 54-1000 kW 70 1000 2.5 12.5 25 2,806,701 32

aaER A-1000/S 70 1000 3 14 20 2,768,736 32

HW77/1500 kW 70 1500 3 12 25 5,854,128 45

GE 77-1.5MW 70 1500 3.5 14 25 5,660,350 43

V82-1.65MW 70 1650 3.5 12 20 6,928,357 48

TWT 82-1.65 70 1650 3 14 25 6,391,646 44

TWT 82-1.65 80 1650 3 14 25 6,934,475 48

TWT 70-1.65 70 1650 3 14 25 4,426,537 30

DeWind D8.2 80 2000 3 13.5 25 6,341,258 36

V80-2.0 MW 80 2000 4 16 25 6,486,139 37

E101-3.0MW 100 3000 2.5 12.5 28-34 10,244,027 39

AW100-3.0MW 100 3000 4 11.7 25 9,804,148 38

V112-3.0MW 125 3000 4 14 25 12,474,35 48

E126-7.5MW 135 7500 2.5 12.5 28-34 16,384,24 25