Optimum Sizing of Hybrid PV/Wind/Battery/Diesel ... - IEEE Xplore

10 downloads 191 Views 2MB Size Report
considers the effect of wind turbine parameters such as rated speed and rated power on electricity cost and compares the performance of various HES.
2012 IEEE International Conference on Power and Energy (PECon), 2-5 December 2012, Kota Kinabalu Sabah, Malaysia

Optimum Sizing of Hybrid PV/Wind/Battery/Diesel System Considering Wind Turbine Parameters Using Genetic Algorithm Abdullrahman A. Al-Shamma’a

Khaled E. Addoweesh

Department of Electrical Engineering King Saud University Riyadh, Saudi Arabia [email protected]

Department of Electrical Engineering King Saud University Riyadh, Saudi Arabia [email protected]

which makes system cost expensive. It is also practical that neither a solar system alone nor a wind system alone can provide a continuous supply due to their periodical and seasonal variations [3]. However, an attractive option for power supply that eliminates to some level most of the disadvantages that are caused due to the stochastic nature of these resources by incorporating two or three complementary resources to form HES in a proper combination [4]. Moreover, integrating different energy resources improves system reliability and decreases system cost. Another advantage of HES is the reduction of energy storage requirements [5] as compared to systems consisting of a single resource. The hybrid Photovoltaic (PV)/wind turbine (WT)/diesel generator (DG) system has been studied comprehensively [610]. However, these studies did not consider the effects of wind turbine parameters such as rated speed and rated power on electricity cost. Additionally, few researchers have compared the performance of various HES at a common location [11]. In order to find the optimal sizing of renewable energy components, five types of WT from different manufacturers with different characteristics were used. Motivated by reducing our dependence on fossil fuel and addressing the shortcoming of previous research into this topic, this paper presents a complete optimization methodology for optimally sizing of HES composes of wind turbines, diesel generator with battery banks as energy storage system using GA. The proposed system intended to supply a remote area located in the northern part of Saudi Arabia. The optimum configuration is not only achieved by selecting combinations with lowest cost, but also by finding a suitable renewable energy fraction that satisfies load demand requirements with zero Loss of Power Supply Probability (LPSP). Five parameters were included in the optimization method; number of PV array, number of wind turbine, wind turbine type, diesel generator rated power and numbers of battery bank.

Abstract—This paper proposes an optimum sizing methodology to optimize the configuration of a hybrid energy system (HES) based on Genetic Algorithm (GA). The proposed methodology considers the effect of wind turbine parameters such as rated speed and rated power on electricity cost and compares the performance of various HES. Furthermore, the relationships between renewable energy fraction and the cost of energy are also given. The proposed method was applied to the analysis of HES which supplies energy for remote village located in the northern part of Saudi Arabia. The decision variables included in the optimization process are the PV array capacity, wind turbine number, battery bank number and diesel generator rated power. Keywords-Hybrid energy system; renewable energg fraction; Genetic algorithm; Cost of energy; Annualized system cost

I.

INTRODUCTION

The rapid reduction of the traditional energy resources of fossil fuel on a worldwide has imposed an imperative search for alternative resources to overcome the worldwide energy consumption increased in the last decade. An additional motivation to reduce our dependence on fossil fuel is the rising indications of the global warming phenomenon which caused a serious climate change. Due to global warming phenomenon, the earth temperature increased by 0.76°C between the years 1850 to 2005 and the warming average rate over the previous 50 years was approximately twice that for the last 100 years [1]. At the earth poles, several areas are warming at a rate of two or three times the global warming average. Even though this small rise in earth temperature may not sound like much it has that huge influence, even a half degree increase in the earth temperature can affect the weather and planet operation. The U.S. Environmental Protection Agency (EPA) notes that the sea level has risen 15 to 20cm in the last 100 years [1] and this rises had a prospectively effects human being populations especially those living in coastal regions and on islands. Solar and wind resources have attracted energy sectors as an alternative options to generate electricity either in gridconnected or stand-alone operation [2]. In contrast, common drawbacks of these options are their natural variations and low availability of these resources and as a result may not match load demand requirements. In addition, using these energy resources independently may result in extensive over-sizing,

978-1-4673-5019-8/12/$31.00 ©2012 IEEE

II.

MODEL OF THE HYBRID SYSTEM COMPONENTS

A HES modeling is a crucial step before any optimal sizing process. The block diagram of proposed HES is shown in Fig. 1. A General methodology for modeling PV, wind turbine, battery bank and diesel generator is described below.

121

2012 IEEE International Conference on Power and Energy (PECon), 2-5 December 2012, Kota Kinabalu Sabah, Malaysia

C. Battery Bank Model The selection of a proper size of the battery bank for HES requires a comprehensive analysis of the battery’s charge and discharge modes, including load profile and output energy of the renewable energy sources. The state of charge (SOC) of battery bank can be calculated from the following equation [9]:

 E (t )  SOC (t ) = SOC (t − 1)(1 − σ ) +  EGA (t ) − L ηbat ηinv  

(4)

Where, SOC(t) and SOC(t-1) are the battery bank state of charge at time t and t-1; σ is hourly self discharging rate; EGA(t) is the total energy generated; EL(t) is the load demand; ηinv and ηbat are the efficiency of inverter and battery. D. Diesel Generator Model Diesel generator is required to supply load when renewable energy is not sufficient. According to Skarstein [16], the fuel cost of the DG can be calculated as follow:

Figure 1. Schematic diagram of HES.

A. PV Module Model Since PV energy is affected by incidence solar radiation on the panel and PV cell temperature, therefore, for estimating hourly output energy generated from PV panel, the solar radiation, ambient temperature and manufacturing data for the PV array are used as model inputs. The hourly output energy of the PV array PPV is given by the following equation [12].

CDG = CF

∑ F (t )

(5)

t =1

Where; F(t) is the hourly fuel consumption and it can be calculated as follow [9]: F ( t ) = APDG ( t ) + BPR (6) Where, PR is the DG rated power, PDG(t) is power generated (kW), CF is the fuel cost per liter and A=0.246 l/kWh and B=0.0845 l/kWh are the fuel curve coefficient [16]. From the above equation, it is noted that, the rated power and generated power influences the fuel consumption of the DG. Therefore, DG should not operate under its minimum setting [6]. Fig. 2 illustrates the cost of energy for a fuel cost of 0.1$/liter. At high loads, the diesel generator operates most efficiently, while the cost per kWh of DG energy becomes very high as the load is decreased towards zero.

 GT  PPV = PR f PV  (1)  1 + α P T c −T c ,STC   GT ,STC     Where, PR is the rated power of the PV; fPV is the PV derating factor; GT ,STC is the incident radiation at standard test conditions; GT is the incident radiation on the tilted PV array; αP is the temperature coefficient of power; Tc is the PV temperature; Tc,STC is the PV temperature under standard test conditions.

(

t =8760

)

B. Wind Turbine Model The wind turbine model is based on rated power (Pr) and three speed characteristics: cut-in (Vc), rated (Vr), and cutoff (Vf). Choosing a suitable model is very important for wind turbine output power simulation. There are several existing models for estimating wind turbine output power, such as linear, quadratic model [13] and model based on Weibull parameter [14]. In this study, the wind turbine output power is approximated by a quadratic model, which can be expressed as:

0.16

C ost of energy($/kW h)

0.14 0.12 0.1 0.08 0.06 0.04

0 V < Vc , V > V f   2 2  (2)  V − Vc PWT (V ) = Pr  2 ............V c ≤ V ≤ V f  2  V r − Vc  Vr ≤ V ≤ V f  1 Since, the wind speeds data are usually available for standard height such as 10m, so that wind speed at tower height must be estimated. One of the most common methods is the power law that can be expressed as [15]:

0.02

0

50 100 150 200 250 Diesel generator output power (kW )

300

Figure 2. Diesel Cost of Energy for CF=$0.1/Liter.

E. System Constraints Model ¾ System Reliability Model In this study, power system reliability is expressed in terms of LPSP [5] which is defined as the probability that an insufficient energy results when the HES is unable to supply the load. This method can be summarized as follows: 9 If load demand (EL(t)) is lower than total renewable energy generated (EGA(t)), then surplus energy is used to charge the battery banks. 9 If EL(t) is higher than EGA(t) and if battery SOC is higher than SOCmin, then deficient energy (EL(t)-

α

Z  Vhub = Vdata  hub  (3)  Z data  Where, Zdata is the height of measurement, Zhub is the height at which wind speed is required, and α is power low exponent.

122

2012 IEEE International Conference on Power and Energy (PECon), 2-5 December 2012, Kota Kinabalu Sabah, Malaysia

The Ccap is given by:

EGA(t)) will be supplied by battery. Otherwise, if battery SOC is equal to SOCmin, DG will be started to supply load as follows: • If EL(t)-EGA(t) is lower than PDGmin, then DG will be operated at its minimum level. • While, if EL(t)-EGA(t) is higher than PDGmin and lower than PR, deficient energy will be supplied by diesel generator. Otherwise, if deficient energy is higher than PR, DG will be operated at PR and the rest of deficient energy will supplied by batteries if stored energy is sufficient. The loss of power supply probability can be defined by [5]:

Ccap = CPV + CWT + CBat + CDG

Where, CPV, CWT, CBat and CDG are the capital cost of PV array, wind turbine, battery bank and DG. The annualized replacement cost is the annualized value of all the replacement costs occurring throughout the lifetime of the project and it can be expressed by:

(

Carep = Crep SSF i, Yrep

∑ PFT

t =0

(7) T The Power Failure Time (PFT) is defined as the time that the load is not satisfied when the energy generated from energy sources is insufficient and the battery bank is exhausted. ¾

Renewable Energy Fraction Model

  ×100  

(8)

Where; EL,DG is load served by diesel generator. Pure renewable system corresponding to REF=100%, while pure diesel system corresponding to REF=0%. So, excluding these boundaries, the remaining values correspond to HES.

The first step of a GA evaluation is to determine whether the chosen system chromosome passes the functional evaluations (LPSP and REF) or not. If the assessment qualifies the chromosome that has a lower COE than the lowest COE value obtained at the previous generation, this chromosome will be selected to be the optimum solution for the minimization problem in this generation. This optimum solution will be replaced by better solutions, if any, produced in subsequent GA generations during the program evolution. After the selection process, the optimum solution will then be subject to the crossover and mutation operations in order to create the next generation population until a pre-specified number of generations is reached, and each new generation is subject to the following inequality constraints:

¾

System Economics Model The economical approach, according to the concept of Cost of Energy (COE), is developed to be the objective function of the system cost analysis in this study and can be expressed as [12]: COE =

ASC

(9) E L , served The annualized system cost (ASC) is composed of the annualized capital cost Cacap, the annualized replacement cost Carep and the annualized maintenance cost Camain. ASC = Cacap + Carep + Camain (10)

min ( PPV , N W T , N Bat , PDG ) ≥ 0 LPSP ≤ LPSPset and REF ≥ REFset

The annualized capital cost of each component is given by:

(

Cacap = Ccap CRF i, Y proj

)

IV.

(11)

TABLE I. THE COSTS AND LIFETIME OF THE SYSTEM COMPONENTS Component Capital Cost Replacement Cost 3.770[$/W] 1.915[$/W] 0.213[$/Wh] 0.857[$/W] 0.714[$/W]

(14)

RESULTS AND DISCUSSION

Genetic Algorithm based on matlab code was developed to determine the optimum sizing of PV/WT/Battery/Diesel system intended to supply a selected remote village called AlSulaymania.

Where, Ccap is the initial capital cost; CRF is the capital recovery factor, Yproj is period lifetime.

Photovoltaic Wind turbine Battery bank Diesel Generator Inverter

(13)

III. SYSTEM OPTIMIZATION USING GENETIC ALGORITHM The flow chart of the proposed GA optimization model is shown in Fig. 4. The input data are solar radiation, wind speed, load profile, REF and specifications of the system devices. The PV array capacity, numbers of battery bank wind turbine and diesel generator rated power are chosen to become the chromosomes of GA. Each chromosome consists of four genes in form of [PPV, NWT, NBat, PDG].

The REF is defined as the fraction of the energy delivered to the load that originated from renewable sources and it can be calculated using the following equation:

 EL, DG REF = 1 −  EL, served 

)

Where, Crep is the replacement cost of the component, SSF is the sink fund factor, Yrep is the component lifetime. The unit price, replacement, O&M cost of the HES components and the lifetime are summarized in Table I. The technical characteristics of the PV, battery banks as well as the technical characteristic of five different types of wind turbines used in this study are listed in Tables II, III and IV, respectively. Hourly data of solar irradiation, wind speed and load demand for a complete year for the selected area are plotted in Fig. 3.

t =T

LPSP =

(12)

Null Null 0.213[$/Wh] 0.857[$/Wh] 0.714[$/W]

123

O&M Cost

Lifetime

Real Interest Rate

1% 3% 1% 0.032[$/kWh] -

25 25 10 10 25

2%[17] -

2012 IEEE International Conference on Power and Energy (PECon), 2-5 December 2012, Kota Kinabalu Sabah, Malaysia

Time (Hour)

Time (Hour)

Figure 3. Meteorological and load demand data for the Village under Study. Figure 3. Diesel Cost of Energy for CF=$0.1/Liter. TABLE II. BATTERY BANK SPECIFICATIONS The COE of diesel system increases Model Nominal capacity Voltage ηBat DOD Surrette-6CS25P TABLE III. BP3135

1156(Ah)

6(V)

PV MODULE SPECIFICATIONS Max. Power Nominal Voltage 135 (W) 12 (V)

80(%)

Efficiency 13 (%)

more rapidly than the COE of HES with an increase in diesel price.

60(%)

B. Impact of REF on COE It can be seen from Fig. 6 that the COE is not linearly dependent on REF, since at high REF approximately from 80100% a little increase in REF will result in high jump in COE value. Therefore, the proposed hybrid energy system with 65% of REF seems to be the suitable renewable energy fraction. Furthermore, it is also depicted from Fig. 6 that the priority sequence for choosing a suitable WT should be the WT3, WT1, WT5 and WT4, then WT2.

NOCT 47 (0C)

WIND TURBINE SPECIFICATION Model Speed Parameters (m/s) Rated Power (kW) Cutin Rated Cutoff FD21-100/12 100 3 12 25 Northern Power 100 3 15 25 DeWind D4/48 600 3 11.5 25 Enercron E33 330 3 13 25 Lagerwey 250 250 3 12 25

TABLE IV. NO. WT1 WT2 WT3 WT4 WT5

Time (Hour)

In this simulation, GA parameters consist of 100 populations and 200 maximum generations. The values of crossover and mutation are 0.7 and 0.25, respectively. These values are determined by trial and error in order to find out optimum solution quickly. LSPS is set to be zero, and REF is set to be 65% and diesel price is assumed to be 0.1$/liter in the optimization. The optimization results for the proposed HES using five wind turbines are summarized in Table V. The proposed HES was also compared with an existing system in the village that uses only diesel generator as shown in Table VI. It can be observed from these results that HES that incorporate wind turbine WT3 gives the lowest COE value compared to the other wind turbines, while WT2 has the highest COE. This shows that selecting a wind turbine with lower rated speed enhances the COE. Therefore, in this study, the priority sequence for choosing a suitable WT should be as follow: WT3, WT1, WT5, WT4 then WT2. Furthermore, it can be also depicted from Table V that wind turbine’s rated power doesn’t affect the COE. From the simulation results shown in table V, it can be seen that the optimum design of HES consists of 1 wind turbine (WT3), 5kW PV arrays, 11 battery banks each of 1156Ah and 296kW diesel generator, which result in a minimum COE of 0.1415$/kWh and the corresponding ASC of $152900. Additionally, take a look into the comparison of the COE between DG system and HES, one familiar reality can be found that the DG system result in a lower COE value compared to the proposed HES. This may be accounted by the relative low price of the fossil fuel in the kingdom, no carbon tax and no encouragement of the usage of clean energy. A. Impact of diesel price on COE The diesel price in the present study has a direct impact on COE value of HES as can be seen from Fig. 5. It is obvious that the proposed HES is always more feasible compared to diesel only system for all diesel prices higher than 0.15$/liter.

Figure 4. Flow Chart of the Optimal Sizing Model using GA.

124

2012 IEEE International Conference on Power and Energy (PECon), 2-5 December 2012, Kota Kinabalu Sabah, Malaysia

TABLE V.

SIZING RESULTS FOR 65% OF REF BASED ON 0.1$/LITER.

Wind Turbine (Type) WT1 WT2 WT3 WT4 WT5

PPV (kW) 115 215 5 195 70

Hybrid Energy System NWT (No) NBatt (Bank) 4 7 4 8 1 11 1 7 2 7

PDG (kW) 297 306 296 305 295

TABLE VI. ANNUALIZED COST OF DIESEL GENERATORS SYSTEM. Capital Cost Fuel Consumption Annualized O&M cost Fuel Prices ($/yr) (Million liter) ($/yr) ($/liter) 40033 0.62743 37863 0.1

Cost of Energy($/KWh)

0.8

0.4

0

0.1

0.2

0.3

0.4 0.5 0.6 Diesel Price($/liter)

0.7

0.8

0.9

1

Figure 5. Impact of Diesel Price on COE of HES and Diesel System. 0.4

Rated Power=600KW, Rated Speed=11.5m/s Rated Power=100KW, Rated Speed=12.0m/s Rated Power=100KW, Rated Speed=15.0m/s Rated Power=250KW, Rated Speed=12.0m/s Rated Power=330KW, Rated Speed=13.0m/s

C os t of E nergy (C O E , $/K W h)

0.35 0.3

0.2 0.15 0.1

30

40

50 60 70 Renewable Energy Fraction (REF, %)

0.1422 0.1635 0.1415 0.1522 0.1425

160000 183900 159200 171200 160300

213570 229680 189170 226440 205520

Annual Fuel Cost ($/yr) 62743

Cost of Energy ($/kWh) 0.12502

Annualized System Cost ($/yr) 140639

The relatively low ASC of HES that incorporates wind turbine and diesel generator augured well for their economic feasibility. Another interesting result seen in Table VII is the reduction in COE caused by the enclosure of batteries to WT/DG system and it is clear that the inclusion of batteries reduces the diesel generator operation and hence reduces the fuel consumption and O&M cost. ASC was the deciding criterion in finding the most economically feasible HES and thus the hybrid PV/WT/Bat/DG and hybrid WT/Bat/DG systems are recommended. Furthermore, a comparison of the ASC between the ten hybrid energy systems investigated in this study revealed that the hybrid PV/WT/Bat/DG, hybrid WT/Bat/DG, and hybrid PV/WT/DG resulted in a lower ASC compared to other energy system. This may be accounted for the abundant wind speed profile in the studied area and the relative higher cost of the PV than that of the wind turbine according to Table I. Therefore, for the load demand in this study, the priority sequence for choosing hybrid energy systems should be PV/WT/Bat/DG, WT/Bat/DG, PV/Bat/DG, PV/WT/Bat, WT/Bat and PV/Bat then hybrid PV/WT system.

0.25

0.05 20

Fuel Consumption (Liters/yr)

It can be also depicted from Table VII that the number of battery banks is found to be higher in HES configurations without diesel generator than HES with diesel generator used as a secondary energy source. The existence of three independent power sources (PV, WT and DG) in the HES increased the system's reliability, while the usage of a limited fossil fuel quantity remarkably diminished the corresponding battery banks number. The diesel system, regardless of having the lowest ACC, is less economical than all the other HES configurations, except for those that include PV without WT and those incorporating renewable sources without diesel generator.

0.2

0

Annualized System Cost (ASC, $/year)

However, ACC fails to take into account O&M and replacement cost of the system during their life time span. Thus, a more accurate assessment of full cost of the HES system, including ACC, O&M cost and replacement cost was achieved via the ASC calculations.

Diesel system Hybrid System Hybrid system Diesel System

0.6

Cost of Energy (COE, $/kWh)

80

90

100

Figure 6. REF impact on cost of energy.

The priority sequence is not affected only by the wind speed conditions but also by wind turbine rated speed. Moreover, WT priority sequence is not highly affect by the wind turbine rated power. C. Comparison between Different System Combinations The optimal sizing method described above can also be applied to design other types of hybrid renewable energy systems, such as those listed in Table VII. Annualized system cost (ASC), annualized capital cost (ACC) and COE were chosen as primary economic evaluation criteria for this study. Optimization analysis has been implemented to these kinds of system types, and the sizing results are illustrated in Table VII. It is interesting to note that, diesel system has the lower ACC compared to the HES. This is due to the expensive nature of renewable energy technology.

Comparing the performance of the hybrid energy systems in term of fuel consumption shows that there is a considerable reduction in diesel consumption attributable to the use of renewable energy resources in the HES. The diesel system consumes more than three times the fuel as compared to all of other selected energy systems. The hybrid PV/WT/Bat/DG and hybrid WT/Bat/DG systems are the lowest systems that consumes diesel among the studied systems.

125

2012 IEEE International Conference on Power and Energy (PECon), 2-5 December 2012, Kota Kinabalu Sabah, Malaysia

TABLE VII.

RESULTS OF THE SYSTEMS INVESTEGATED IN THIS STUDY.

HES

PV/WT/Bat/DG PV/Bat/DG WT/Bat/DG PV/WT/ DG PV/WT/Bat WT/Bat PV/Bat PV/WT WT/DG DG

NWT (No) 1 0 1 1 2 9 0 17 1 0

HES Components PPV NBat PDG (kW) (Bank) (kW) 5 11 296 505 54 322 0 14 299 45 0 309 510 170 0 0 314 0 4045 227 0 3515 0 0 0 0 330 0 0 2×345

Cost of Energy (COE, $/kWh)

Annualized Capital Cost (ACC, $/kWh)

Annualized System Cost (ASC, $/yr)

Fuel Consumption (liter/yr)

CO2 (kg/yr)

0.141524 0.209974 0.143124 0.146680 0.341986 0.738999 0.871900 1.677734 0.167926 0.180800

96439 148290 96513 101400 287940 645010 885240 1699500 93637 40033

159200 255620 179720 188150 384700 831300 1043400 1736300 188900 203382

189170 195000 187000 231000 0 0 0 0 257430 627430

286000 282000 284000 316000 0 0 0 0 353510 827060

[2] P. Balachandra, N.H. Ravindranath Deepak Paramashivan Kaundinya, "Grid-connected versus stand-alone energy systems for decentralized power—A review of literature," Renewable and Sustainable Energy Reviews, vol. 13, p. 2041–2050, 2009.

V. CONCLUSIONS A methodology of optimum sizing of a standalone HES consisting of PV arrays, wind turbines, battery banks and diesel generator using GA is presented in this paper. This optimization methodology takes in consideration the effect of wind turbine parameters such as rated speed and rated speed. Moreover, the resented method was applied to ten configurations of hybrid renewable energy systems.

[3] Chengzhi Lou, Zhongshi Li, Lin Lu, Hongxing Yang Wei Zhou, "Current status of research on optimum sizing of stand-alone hybrid solar–wind power generation systems," Applied Energy, vol. 87, p. 380– 389, 2010. [4] Kolokotsa D., Potirakis A., Kalaitzakis K. Koutroulis E., "A Methodology for optimal sizing of stand-alone photovoltaic/windgenerator systems using genetic algorithms," Solar Energy, vol. 80, p. 1072–1088, 2006.

The optimization sizing of the HES was applied to a practical case in Saudi Arabia, which is a remote village. According to the results related to the case studied in this paper, it can be concluded that:

[5] Lu L, Zhou W Yang HX, "A novel optimization sizing model for hybrid solar–wind power generation system," Solar energy, vol. 81, p. 76–84, 2007. [6] Nayar C.V. Ashari M., "An optimum dispach strategey using set points for a photovoltaic (PV)-diesel-battery hybrid power system," Solar Energy, vol. 66, pp. 1-9, 1999.

• The optimum sizing depends on renewable energy fraction (REF) and system reliability (LPSP). Hence, to meet load demand at high reliability and at high REF, there is an extensive increase in the system sizing. • At high REF value (80-100%), the COE rises sharply for a little increase in REF. • The wind turbine rated speed has a significant effect on COE and a wind turbine with the lowest rated speed found to be the most economical one. On the other hand, wind turbine rated power doesn’t affect the COE. • The battery bank has an important effect on COE and the enclosure of batteries to the HES reduces the required DG requirements and hence reduces the fuel consumption. As a result O&M cost and COE are reduced. • For the load demand in this study, the most economical systems are the hybrid PV/WT/Bat/DG, hybrid WT/Bat/DG, and hybrid PV/WT/DG as they result in a lower COE compared to the other hybrid energy systems. • The proposed HES is always feasible compared to diesel system for all diesel price higher than 0.15$/liter. • The COE of diesel system increases more rapidly than the COE of HES with an increase in diesel price.

[7] Hayashia D., Yonaa A., Urasakia N., Funabashib T. Senjyua T., "Optimal configuration of power generating systems in isolated island with renewable energy," Renewable Energy, vol. 32, p. 1917–1933, 2007. [8] Al-Hadhrami L.M. Rehman S., "Study of a solar PVedieselebattery hybrid power system for a remotely located," Energy, vol. 35 , pp. 49864995, 2010. [9] Zhang L., Barakat G. Belfkira R., "Optimal sizing study of hybrid wind/PV/diesel power generation unit," Solar Energy, vol. 36, pp. 100110, 2011. [10] M.A. Elhadidy S.M. Shaahid, "Technical and economic assessment of grid-independent hybrid photovoltaic-diesel-battery power systems for commericial loads in desert environments," Renewable and Sustainable Energy Review, vol. 11, no. 8, pp. 1794-1810, 2007. [11] Campbell H., Sanguinetti C. Hessami M.A, "A feasibility study of hybrid wind power systems for remote communities," Energy Policy, vol. 39, pp. 877-886, 2011. [12] National Renewable Energy Laboratory (NREL). Homer-Analysis of micropower system options. [Online]. http://analysis.nrel.gov/homer/default.asp [13] Pallabazzer R., "Evaluation of wind generator potential," Solar Energy, vol. 55, pp. 49-59, 1995. [14] Johnson G.L., Wind energy systems. USA: Prentice Hall, 1985.

ACKNOWLEDGEMENTS

[15] Wang Li Yeh Tai-Her, "A study on generator capacity for wind turbines under various tower heights and rated wind speeds using weibull distribution," IEEE Transaction On Energy Conversion, vol. 23, pp. 592602, 2008.

This work was financially supported by the National Plan for Science and Technology (NPST) program, King Saud University; Project Number: 09 ENE 741-02 and KACST student grant: A-S-11-0731.

[16] Ulhen K. Skarstein O., "Design considerations with respect to long-term Diesel saving in wind/Diesel plants," Wind Engineering;, vol. 13, pp. 7287, 1989.

REFERENCES

[17] TRADINGECONOMICS. (2012, January) Saudi Arabia Interest Rate. [Online]. www.TRADINGECONOMICS.COM

[1] Whole Global Greenhouse Warming Internet site. Global Greenhouse Warming. [Online]. http://www.global-greenhouse-warming.com.

126