Predictive model and assessment of the potential for wind and solar power in Rayak region, Lebanon

Abstract

With the increasing consumption of fossil fuel, reducing greenhouse gas (GHG) emissions has become a serious issue that has attracted worldwide attention. Therefore, Lebanon is currently interested in utilizing renewable energy technologies to reduce energy dependence on oil reserves and GHG emissions. The present study is focused on solar and wind power potential and the economic viability of wind/solar systems for the Rayak region in Lebanon for the first time. The input data sources for the study including the Meteorological data for wind speed and NASA database for solar energy. In the assessment of wind energy, a two-parameter Weibull distribution function was used to analyze the characteristics of wind speed. Yearly and seasonal Weibull parameters were calculated for 10 m height using the Maximum likelihood method. In addition, yearly and seasonal wind power density values were calculated. The results showed that the mean annual wind speed and wind power density values were 5.884 m/s and 124.534 W/m², respectively, during the investigation period. It can be concluded that the value of wind power density at the region was classified as marginal wind power potential and high-scale wind turbines can be used to gather wind energy potential in the region. Furthermore, predicting wind speed depends on various atmospheric factors and random variables. Therefore, 63 ANN models are developed by varying the meteorological parameters to predict the daily wind speed in the selected region. All the models with various combinations are validated and the performances of the models are analyzed using root mean squared error. The results demonstrated that the most relevant input variables for predicting the daily wind speed were found to be temperature, pressure, and relative humidity. In the assessment of wind energy, average monthly global solar radiation data were evaluated. It is found that the annual global solar radiation was 1877.41 kWh/m², which indicates the selected region has high solar resources and is categorized as an excellent potential class. Moreover, this study provides a comprehensive and integrated feasibility analysis of 100 MW grid-connected wind and solar projects economic projects that can be developed in the country to reduce the electricity crisis and GHG emissions. Several different economic and financial indicators were calculated. The results indicate that the wind farm is a more economical option than the solar plant because of the higher values of NPV, BCR, ALCS, and IRR as well as the lower values of EB, SB, and LCOE.

Introduction

The utilization of solar and wind energy is expected to increase more than any other energy source by the middle of this century (Owusu and Asumadu-Sarkodie 2016; Osinowo et al. 2017). Solar and wind energy are clean and renewable sources of energy (Prasad et al. 2017; Kassem 2018). Therefore, wind and solar energy are alternative energy sources and are being progressively adapted in numerous applications(Zendehboudi et al. 2018; Chang and Starcherl 2019). Many studies have been investigated by solar and wind energy generated electricity potential in many countries worldwide to generate power. For example, Zhou et al. (2011) evaluated the onshore wind energy potential in Juangsu province, china based on meteorological data from 1979 to 2008. The results indicated that the annual power production from wind found to be 146,336 GWh in the selected location. Ahmed (2019) presented the technical assessment of a two wind farm of 300 MW-installed capacity at chosen regions in Egypt. It is found that the annual power production from the proposed wind farms is found to be 1130 GWh and the expected cost was varied from 1.96€ to 2.09 € cents/kWh. Dabar et al. (2019) assessed the wind potential in the Republic of Djibouti based on 3 years by meteorological stations at eight locations. In addition, they examined the feasibility of three wind farms of 275 MW installed capacity at GaliMa-aba, Ghoubbet, and Bada Wein. The results showed that the proposed wind farms would produce 1073 GWh/year of electricity and the expected cost was varied from 7.03 to 9.67 US.$ cent/kWh. Chadee et al. (2018) conducted the technical and economic assessments to estimate the expected cost of electricity generation from wind in Trinidad and Tobago islands based on long-term measurements of wind speed. The results found that electricity cost for the large wind turbines, 1500 kW and greater, were less than the current subsidized residential tariff of the US/$0.06/kWh. Parikh et al. (2019) determined the variation in electricity solar potential by utilizing different types of solar panels. The results concluded that the maximum value of power generated from the proposed farm was estimated to be 173.6 MW. Rehman et al. (2017) presented the feasibility of 10 MW grid-connected PV plants at 44 sites in Saudi Arabia. It is found that the best location for the future installation of a PV plant was found to be Bisha due to the highest solar radiation intensity and longer sunshine duration. Adaramola (2014) studied the feasibility of the PV grid-tied energy system in Jos, Nigeria using HOMER software. The results showed that the system could be produced energy of 331.536 GWh/year with a capacity factor of 40.4% from solar energy.

Lebanon, which is located in the Eastern Mediterranean, covers a total area of 10,452 km², with a coastline of about 220 km from north to south. Lebanon borders Syria to the north and east, and Israel to the south. Geographically, Lebanon consists of a narrow coastal strip that extends parallel to the Mediterranean Sea and two mountain ranges, namely Mount Lebanon and the East Lebanon Mountain Range or Qalamoun Mountains, running from north to south. The fertile Bekaa Valley is located between the two ranges. Lebanon usually has a Mediterranean climate, with cold and rainy winters from the month of December until mid-March, and hot and dry summers that begin in June and ends in September. Average temperatures across the country vary from 20 °C on the coast to 17 °C in the Bekaa Valley and less than 10 °C in the mountains.

Electricity of Lebanon

Lebanon is among the few countries in the world without electricity 24/7, and suffers from a large energy deficit. The energy production depends on fuel oil, diesel oil, and other imports, in which results in electricity blackout periods that range from 4 to 14 h in all Lebanese regions and sometimes more. Currently, the electrical energy in Lebanon is generated by oil products, hydropower, and importation. The power plants in Lebanon are classified as thermal and hydroelectric. In fact, Lebanon has 13 power plants with a total capacity of 3016 MW distributed as 2764 MW thermal plants and 252.6 MW hydropower plants. Statistics indicate that 97% of the total energy was generated by thermal plants in 2015 and hydropower had a share of only 2.6% in that year, as shown in Fig. 1. The electricity crisis is one of the most significant issues affecting the daily lives of citizens, shop owners, and small businesses which has been at the forefront of daily life in Lebanon for years. Lebanon is experiencing a real crisis because of the power cuts that have continued for many years without any proven solutions. The electricity service is becoming worse in Lebanon. According to the Lebanon News, the number of hours during which the electricity supply is disrupted in all cities of the country is increasing day by day. Lebanon is not a country isolated from the world and the electricity sector crisis cannot be separated from the conditions of the global economy. International oil prices play a major role in increasing the deficit in the EDL. The higher the price of oil globally, the more the EDL losses will increase, because 85% of the country’s energy production is dependent on this oil. In addition, the causes of the crisis in the electrical sector in Lebanon are the lack of alternatives to petroleum and the low production capacity (the Electricity Corporation produces 1500 mW, while the demand reaches three thousand megawatts, which requires the construction of new factories). Therefore, the increase in the population and subsequent demand for energy have enhanced the significance of renewable energy as an alternative source in recent years.

Fig. 1

Electricity production from oil and hydroelectrically sources (World Bank 2019)

Currently, special diesel generators (DG) distributed throughout the country are used to meet the energy consumption needs of the population. During periods of power outages, the central public directorates in each region take responsibility and provide users with electrical energy. Subscriptions vary according to the needs of each consumer. While the majority of apartments may settle for the cheapest offer, general managers provide 1.1 kVA, and other citizens residing in larger homes with a higher standard of living tend to subscribe to more expensive offers that provide 2.2–4.4 kVA. The resulting fees increase in accordance with the subscription. It is worth noting that most of the time. On the other hand, industries, hospitals, offices, and centers tend to install their own generators according to their energy consumption. This solution provides them with complete control over electricity production during a power outage and helps reduce the fees generated, or at least the ability to manage them. Installing DGs is an alternative solution that not only causes massive air pollution, but is very uncomfortable due to the industry's medium regulation. Moreover, it does not provide flexibility to the user, meaning that a fixed amount of money is imposed on the population independently of the amount of energy consumed. The resulting fees are highly influenced by fuel prices and blackout hours. Recently, Lebanese municipalities began charging a fixed monthly tariff for each subscription to try to contain the chaos and price differentials between different regions of the country.

Renewable energy in Lebanon

Lebanon is located in the Eastern Mediterranean with a surface area of 10,452 km² and a coastline of 220 km. The population in the country is estimated to be 4 million. As the Lebanese electricity system is mainly based on thermal energy, large volumes of dangerous gases are emitted into the atmosphere on a daily basis. According to Chedid et al. (2001), the energy sector in Lebanon was found to contribute 85% of all CO₂ emissions and 96% of all SO₂ emissions. Therefore, renewable energy projects account for a worldwide decrease in CO₂ levels in several countries. Renewable energies are offered by natural sources that can be persistently and sustainably replenished. In general, PV, wind and hydropower are the most interesting resources to explore in Lebanon. As mentioned previously, around 2% of the total energy in the country is produced by hydropower stations. According to the UNDP, the wind potential is estimated to be 1500 MW in Lebanon. Moreover, according to Abdeladim et al. (2018), Lebanon has high levels of sunshine hours and the daily average solar insolation is about 4.8 kWh/m². Numerous studies have investigated the potential of renewable energy in terms of solar energy and wind energy in Lebanon. For instance, Al Zohbi et al. (2015) investigated the characteristics of wind speed for five locations in Lebanon. They found that wind power could reduce the electricity demand in Lebanon. Zohbi et al. (2016) evaluated the performance of a hybrid wind-hydro power plant in two dams in Lebanon to find the best dam to generate energy by wind power at night. The authors concluded that a combination of wind energy with a pumped hydro storage system could be an ideal solution to solve Lebanon's electricity crisis. Kassem et al. (2019a, b, c, d, e) investigated the wind potential in three coastal regions (Beirut, Sidon, and Tripoli) in Lebanon. The authors found that vertical axis wind turbines have the potential to outperform horizontal axis wind turbines in urban environments. Gökçekuş et al. (2019) analyzed the wind speed characteristics and wind energy potential at eight selected locations in Northern Lebanon. They concluded that small-scale wind turbine use could be suitable for generating electricity in the studied regions. Kassem et al. (2019a, b, c, d, e) investigated the wind potential at certain selected regions in Lebanon using the Two-Parameter Weibull distribution function. The results indicated that the wind speeds at the selected regions were within the range of 2.627 m/s and 3.56 m/s. Furthermore, the wind speed densities varied between 14.634 and 25.280 W/m², which is classified as poor wind power. Tannous et al. (2018) investigated and compared traditional grid-connected and solar stand-alone systems for two street lighting technologies in Lebanon. The results showed that the solar system had a lesser overall environmental impact than the traditional system. Ibarra-Berastegi et al. (2019) evaluated the Lebanese offshore-wind-energy potential. They found that wind power density along the shore of the northern coastal area exhibited the highest potential and reached r values of around 400 W/m² in the winter. Elkhoury et al. (2010) investigated the feasibility of wind turbine farms for generating electricity in four sites in Lebanon. The results indicated that Zahleh and Beirut have high wind energy potential. Nakad et al. (2012) studied the techno-economic feasibility of a wind farm in the Zahleh region taking into consideration the land area and the appropriate turbines.

Importance of the study

Based on the literature review, it is evident that there is a clear lack of utilization of wind energy potential as power generation sources in Lebanon. To the best of the author’s knowledge, no study has investigated the wind potential in the city of Rayak, Lebanon. Moreover, no study has focused on the feasibility of a 100 MW grid-connected wind/PV in the selected city to find the most economic project in the region. According to electricity of Lebanon (Électricité du Liban), the total power for Beqaa Valley is estimated to be 300 MWh. This study aims to reduce the electricity consumption generated by diesel fuel by 33%. Therefore, the aim of this paper is to analyze the wind speed characteristics for Rayak city, Bekaa, Lebanon based on data measured at height of 10 m Meteorology department. The hourly data cover the period of 1996–2006. In addition, to identify the meteorological parameters that influenced the prediction of wind speed, 63 ANN models were developed. Moreover, the monthly NASA database is used to assess the solar potential in the selected region. Furthermore, the use of wind or solar energy to power such systems may solve the electricity crises in the country and reduce GHG emissions. In this regard, the feasibility of a 100 MW grid-connected wind/PV project is presented. For this purpose, RETScreen Expert software is used.

Materials and methods

In this section, the potential of wind and solar energy is discussed. To investigate the wind potential, the wind speed characteristics, which were measured at a height of 10 m at Rayak region, are statistically analyzed. The Weibull distribution function is used to determine the wind power density in the selected locations. Furthermore, the power-law method is utilized to estimate the wind speed at various hub heights. Moreover, in general, there are several variables that can influence the potential wind resource in a specific region. In addition, the prediction of wind speed is important for estimating the wind power density in the region; thus, determining the most significant parameters is an important task to achieve accurate predictions of wind speed. In this study, an Artificial Neural Network (ANN) is used to identify the most relevant parameters for the prediction of the daily wind speed of the selected region. Moreover, the solar energy potential in the chosen location is investigated based on NASA's average monthly global solar radiation data. In addition, for the future installation of a grid-connected 100 MW wind and solar system, RETScreen software is used to present technical, environmental, and economic aspects. Figure 2 illustrates the analysis procedure of the current study.

Fig. 2

Flowchart for analysis steps of the study

Location selection

Figure 3 shows a regional map of Lebanon. The location chosen for the present study is in the west of the country. Rayak is a city located in the Zahleh District, one of the districts of the Bekaa Governorate. The city of Rayak rises 930 m above sea level and extends over an area of 992 hectares (9.92 km²). The city contains a military airport. In general, the Beqaa valley is named due to its location between the eastern and western chains, moving northeast to southwest, and creates a 30-km stretch over an area equal to one-third of Lebanon. Its average height is no more than 900 m, and its highest point is in Baalbek, and it is considered the area of water that divide the rivers of Orontes and Litani. In administrative terms, Beqaa Valley is divided into three sections: North Beqaa, Middle Beqaa and West Beqaa. In the Middle Beqaa region, the humidity is high in winter, with Riyak station recording 82% in January. In the summer, it is about 48%. The relative humidity trajectory is quite different in the coastal areas, as the data shows it that is at its highest in summertime and drops in the winter in the coastal areas. For example, the humidity in the Beirut area reaches its highest value in summer of approximately 73% in July and August and the lowest in November at 68%. In general, these areas do not experience an average of less than 68% in almost all months.

Fig. 3

Lebanon map

Wind data analysis procedure

Weibull probability density function

The Weibull distribution is the most commonly used for analyzing the wind speed ($v$) characteristics in a specific region. The maximum likelihood method (MLM) is widely used for estimating the Weibull parameters. The probability density ($f\left( v \right)$) and cumulative distribution ($F\left( v \right)$) functions for two-parameter Weibull distribution are expressed in Eqs. (1) and (2) (Khan et al. 2018; Alayat et al. 2018; Kassem et al. 2019c). In addition, the mean velocity of the two-parameter Weibull distribution ($\overline{v}$) can be calculated using Eq. (3):

$$ f\left( v \right) = \left( \frac{k}{c} \right)\left( \frac{v}{c} \right)^{k - 1} {\text{e}}^{{ - \left( \frac{v}{c} \right)^{k} }} , $$

(1)

$$ F\left( v \right) = 1 - \exp \left[ { - \left( \frac{v}{c} \right)^{k} } \right], $$

(2)

$$ \overline{v} = c\Gamma \left( {1 + \frac{1}{k}} \right), $$

(3)

where c is the scale parameter in m/s and k is the shape factor of distribution.

The Maximum likelihood method is a mathematical technique used to estimate the Weibull parameters through a numerical iteration. The Weibull parameter k and c values are calculated by the following equations (Kassem et al. 2019b, d):

$$ k = \left( {\frac{{\mathop \sum \nolimits_{1}^{n} v_{i}^{k} \ln \left( {v_{i} } \right)}}{{\mathop \sum \nolimits_{1}^{n} v_{i}^{k} }} - \frac{{\mathop \sum \nolimits_{1}^{n} \ln \left( {v_{i} } \right)}}{n}} \right)^{ - 1} , $$

(4)

$$ c = \left( {\frac{1}{n}\mathop \sum \limits_{1}^{n} v_{i}^{k} } \right)^{1/k} . $$

(5)

Moreover, the probable wind speed ($V_{{{\text{mp}}}}$) and maximum energy carrying wind speed ($V_{{{\text{maxE}}}}$) are essential wind speed indicators, which can be calculated using Weibull parameters as given below (Fazelpour et al. 2017; Pishgar-Komleh et al. 2015):

$$ V_{{{\text{mp}}}} = c\left( {1 - \frac{1}{k}} \right)^{1/k} , $$

(6)

$$ V_{{{\text{maxE}}}} = c\left( {1 + \frac{2}{k}} \right)^{1/k} . $$

(7)

Wind power density

The theoretically available kinetic energy that wind possesses at a certain location can be expressed as the mean wind power density (WPD). In other words, it is the maximum available wind power at each unit area. The mathematical expression for wind power density is given with the following relations (Mohammadi et al. 2017; Kassem et al. 2019c):

$$ \frac{P}{A} = \frac{1}{2}\rho v^{3} , $$

(8)

$$ \frac{P}{A} = \frac{1}{2}\rho v^{3} f\left( v \right). $$

(9)

Furthermore, WPD can be determined using Weibull parameters as follow (Keyhani et al. 2010):

$$ \left( \frac{P}{A} \right)_{W} = \mathop \smallint \limits_{0}^{\infty } \frac{1}{2}\rho v^{3} f\left( v \right){\text{d}}v = \frac{1}{2}\rho c^{3} \Gamma \left( {1 + \frac{3}{k}} \right). $$

(10)

Moreover, the mean wind power density can be estimated using Eq. (11) (Irwanto et al. 2014):

$$ \frac{{\overline{P}}}{A} = \frac{1}{2}\rho \overline{v}^{3} , $$

(11)

where P is the wind power density in W, ${ }\overline{P}$ is the mean wind power density in W, A is the swept area in m², ρ is the air density, $f\left( v \right)$ is the probability density function (PDF), and $\overline{v }$ is the mean wind speed in m/s.

The monthly air density can be calculated using the following equations:

$$ \rho = \frac{{P_{{{\text{av}}}} }}{{R_{t} \times T_{{{\text{av}}}} }} , $$

(12)

where $P_{{{\text{av}}}}$ is the average air pressure in Pa, $R_{t}$ is the specific gas constant (287 J/kg K) for air and $T_{{{\text{av}}}}$ is the average air temperature in K.

In general, pressure, temperature and relative humidity are important factors affecting the air density, i.e., the air density is reduced by decreased air pressure, increased temperature and humidity (Danook et al. 2019).

The percentage errors in estimating the wind power density are calculated as follows:

$$ {\text{Error}} = \left| {\frac{{\left( \frac{P}{A} \right)_{{\text{W}}} - \left( \frac{P}{A} \right)_{{\text{A}}} }}{{\left( \frac{P}{A} \right)_{{\text{W}}} }}} \right| \times 100\% { }, $$

(13)

where $\left( \frac{P}{A} \right)_{{\text{A}}}$ is the actual wind power density and $\left( \frac{P}{A} \right)_{{\text{W}}}$ is the Weibull wind power density.

Wind speed and air density extrapolation

The power law model is widely used to estimate the wind speed (v) at different wind turbine hub heights (z) (Gul et al. 2019; Shu et al. 2015). It is expressed as

$$ \frac{v}{{v_{10} }} = \left( {\frac{z}{{z_{10} }}} \right)^{\alpha } , $$

(14)

where $v_{10}$ is the wind speed at original height $z_{10}$, and $\alpha$ is the surface roughness coefficient (Eq. 14):

$$ \alpha = \frac{{0.37 - 0.088\ln \left( {v_{10} } \right)}}{{1 - 0.088\ln \left( {z_{10} /10} \right)}}. $$

(15)

Moreover, at sea level, the air density has an approximate value of 1.225 kg/m³. However, the air temperature at a specific region is not constant, because the site is an open space consisting of a mixture of dry air and a small amount of water vapor. The time varying air density at the considered heights is estimated using the below equation (Olaofe and Folly 2013):

$$ \rho \left( h \right) = \frac{{P_{av} }}{{RT_{av} }}\exp \left( { - \frac{gh}{{RT}}} \right), $$

(16)

where $\rho \left( h \right)$ is the time varying air density as a function of height in kg/m³, $P_{{{\text{av}}}}$ is the average atmospheric pressure in hPa, R is the molar gas constant [287.05 J/(K mol)], $T_{{{\text{av}}}}$ is the air temperature (K), g is the gravitational constant (9.81 m/s²), and h is the considered height(s) above sea level.

Influence of ambient conditions on wind speed measurement

In general, there are several variables that can influence the wind speed measurement as discussed in Guerrero-Villar et al. (2019) and Danook et al. (2019). Guerrero-Villar et al. (2019) experimentally analyze the effect of temperature, humidity and atmospheric pressure on the cup anemometer behavior. Danook et al. (2019) studied the impact of relative humidity on the air density. The results indicated that humid air implies a lower density resulting in lower power from a wind turbine. Therefore, the aim of this section is to find the input parameter has the greatest influence in terms of affecting the prediction of wind speed in the selected region using an artificial neural network (ANN). To achieve this, seven parameters, namely maximum temperature, minimum temperature, maximum humidity, minimum humidity, Altimeter Pressure Settings, Atmospheric pressure at aerodrome elevation and number of days are considered as input parameters in the ANN models. To check the prediction accuracy using the identified parameters, 63 ANN models are developed.

Artificial neural network (ANN)

Artificial neural networks (ANNs) are known as potent nonlinear mathematical paradigms can be utilized to model the behavior of complex systems (Laqui et al. 2019). The ANN aims to estimate the output value from input values by some internal calculations. ANNS have been adopted in several industrial and scientific fields (Kassem et al. 2019e). Feed-forward ANNs are the most practical types of artificial intelligence paradigms in approximating functions. ANN approaches consist of three types of layers called input, hidden and output layers. In the present study, nine parameters were selected for predicting daily wind speed, as shown in Table 1. Moreover, the structure of ANN used in this paper is shown in Fig. 4. In this study, the number of epochs and performance goal were 100,000 and 0.001, respectively. In addition, the number of hidden layers varied between 1 and 10, while the number of neurons varied between 5 and 50 s.

Table 1 Input and output variables

Fig. 4

Structure of ANN

Training and testing

The Levenberg–Marquardt (LM) algorithm is one of the most widely used and validated back-propagation algorithms. In general, the activation function for the neurons can be linear or non-linear. The logistic-sigmoid (logsig) and tangent-sigmoid (tansig) are used as activation functions whose outputs lie between 0 and 1 and are defined as

$$ \log {\text{sig}} = \frac{1}{{1 + {\text{e}}^{ - x} }} , $$

(17)

$$ {\text{tansig}} = \frac{{{\text{e}}^{x} - {\text{e}}^{ - x} }}{{{\text{e}}^{x} + {\text{e}}^{ - x} }}. $$

(18)

The key step for developing an ANN is the training procedure, where the weights and biases are adjusted to minimize the difference between the output of the ANN and the actual value. To find the best performance for the ANN trained model, the mean squared error (MSE) is used. Equation (19) is used to normalize the data in the range of 0–1 and Eq. (20) is used to return the data to the original values after the simulation:

$$ x_{n} = \frac{{x_{{{\text{actual}}}} - x_{\min } }}{{x_{\max } - x_{\min } }}, $$

(19)

$$ x_{{{\text{actual}}}} = x_{n} \left( {x_{\max } - x_{\min } } \right) + x_{\min } . $$

(20)

Wind speed prediction with selected inputs

Figure 5 shows the methodology used in this study. In the present study, Tmin, Tmax, Hmin, Hmax, QFE and QNH are the inputs of the ANN models and WS is the output of the models. The optimum number of hidden layers and neurons are determined by a trial and error approach to obtain the best performance of the ANN model. In this research, 80% and 20% sets of training data and testing data were randomly selected, respectively. The testing data do not affect training and provide an independent measure of network performance during and after training. The developed models with different combination used to predict the wind speed of the selected regions are listed in Table 2.

Fig. 5

Proposed ANN model for predicting the LUC and CP

Table 2 ANN models developed in this study for predicting wind speed

Appraisal of the developed models

The developed ANN models were evaluated comprehensively to predict the monthly rainfall. The following statistical indicators were employed: mean squared error (MSE), root mean squared error (RMSE) and mean relative error (MAE):

$$ R^{2} = 1 - \frac{{\mathop \sum \nolimits_{i = 1}^{n} \left( {a_{a,i} - a_{p,i} } \right)^{2} }}{{\mathop \sum \nolimits_{i = 1}^{n} \left( {a_{p,i} - a_{{a,{\text{ave}}}} } \right)^{2} }} , $$

(21)

$$ {\text{MSE}} = \frac{1}{n}\mathop \sum \limits_{i = 1}^{n} \left( {a_{a,i} - a_{p,i} } \right)^{2} , $$

(22)

$$ {\text{RMSE}} = \sqrt {\frac{1}{n}\mathop \sum \limits_{i = 1}^{n} \left( {a_{a,i} - a_{p,i} } \right)^{2} ,} $$

(23)

where $n$ is the number of data, $a_{p,i}$ is the predicted values, $a_{a,i}$ is the actual values, $a_{{\text{a,ave}}}$ is the average actual values and i is the number of input variables.

Description and design of the wind or PV project

Description and design of the wind power project

It is not possible to convert all the power produced by the wind turbine into electrical power. Power conditioning, a transmission mechanism, and rotor are significant parameters that affect the electrical power (Alsaad 2013). The theoretical power of the wind, $P$ is given as

$$ P = \frac{1}{2}\rho Av^{3} = \frac{1}{2}\rho \left( {\frac{\pi }{4}d^{2} } \right)v^{3} , $$

(24)

where ρ is the air density (ρ = 1.25 kg/m³), d is the rotor diameter in m, and v is the wind speed of the region. Using Eq. (24), the power produced from the wind becomes

$$ P = \frac{\pi }{8}\rho d^{2} v^{3} = \frac{1}{2}d^{2} v^{3} . $$

(25)

According to Alsaad (2013), the efficiency value for the power conditioning is about 90%, the transmission system is 40%, and the rotor ranges from 35 to 40%. Therefore, the total efficiency ranges from 13 to 15%, which is the result of multiplying all efficiencies of the power conditioning, transmission mechanism, and rotor.

Thus, a coefficient called Hugh’s coefficient ($C_{H}$) must be included in Eq. (25) to determine the actual wind power. As a result, the actual wind power can be estimated using Eq. (26) (Alsaad 2013):

$$ P = C_{H} d^{2} v^{3} . $$

(26)

In general, wind speed is the most important factor that affects the amount of electricity produced from the wind turbine. Therefore, designing a wind farm in a specific region depends on the wind speed and wind turbine characteristics. This theoretical design is used to obtain the nominal capital cost, which can help in conducting a feasibility study for the present research. Moreover, the amount of the time that the wind farm project is able to generate electricity over an specific period of time investigated divided by the total amount of available time during the period is called the availability factor (AF). For the wind turbine, the availability is determined by deducting the time intervals where speed of the wind is lower than cut-in or higher than the cut-out speeds of the wind turbine. The AF is defined as (Alsaad 2013):

$$ {\text{AF}} = 1 - \frac{n}{{N{ }}}, $$

(27)

where n is the number of months when the wind speed is lower than the specified cut-in speed and N is the total number of months during the investigation period.

In this study, the selected wind turbine, AN BONUS 1 MW—70 m with a capacity of 1000 kW is selected after an overall comparison of various wind turbine types. The selected wind turbine that satisfies the estimated annual energy for the selected location is shown in Table 3.

Table 3 Specifications of wind turbine

Description and design of the solar PV project

Lebanon has to make an important decision regarding its energy infrastructure in the coming years, especially in the power sector. In addition, Lebanon has tremendous scope for generating solar energy. Based on the global solar atlas map, it is found that the annual average of GHI of Lebanon varies from 1680 to 2118 kWh/m². The highest global horizontal irradiation is in the western part of Lebanon, which ranges from 5.4 to 5.8 kWh/m²/day. Furthermore, it is noticed that the annual DNI values of the western part of Lebanon are within the range of 6.4–7.2 kWh/m²/day. These results indicate that Lebanon has huge solar energy potential for assessing the energy generation for PV/flat-plate photovoltaic, CSP and CPV technologies. In addition, air temperature (TEMP) is considered as the second important parameter for evaluating the performance of solar PV systems. Based on the global solar atlas map, it is observed that the air temperature values vary from 5.3 to 21.9 °C. It can be concluded that Lebanon has huge solar potential and it can be used to generate power, i.e., the PV output power is within the range of 3.99–5.30 kWh/kWp according to the global solar atlas map. In general, the global solar atlas map is the most reliable source of data currently available that is used to estimate the solar potential in the specific region. It is sufficient for the first step in the decision-making process in terms of the proposed solar PV project in the country. Furthermore, in solar PV project development stage, meteorological data are required for the technical design, engineering, financing, risk assessment, and due-diligence.

In this section, NASA average monthly global solar data are used to evaluate the solar potential for the selected region in Lebanon. In addition, the technical specifications of solar PV system, design methodology, and parameters are discussed in detail in Kassem et al. (2020).

Power generating factor

The power generation factor (PGF) is an important factor that must be considered in the sizing of solar photovoltaic technology based on the total watt peak rating of the system. The PGF is calculated using the below equation:

$$ {\text{PGE}} = \frac{{{\text{Solar irradiance}} \times {\text{Sunshine hours}}}}{{\text{ Standard test condition irradiance }}}. $$

(28)

Energy demand

The total power and energy consumption of all loads that will be supplied by the system for the selected location are essential factors needed to design a solar PV system.

Solar PV energy required

The energy that must be generated by the PV module is the total Watt-hour per day needed from the PV modules and it is calculated using Eq. (29):

$$ {\text{The energy required from PV modules }} = {\text{Peak energy requirement}} \times {\text{Energy lost in the system}}. $$

(29)

PV module sizing

To calculate the size of the photovoltaic modules that are required, it is first necessary to estimate the total Watt-peak rating needed for the PV modules using the formula below:

$$ {\text{Total Watt peak rating}} = \frac{{\text{Solar PV energy required}}}{{\text{ panel generation factor }}} , $$

(30)

$$ {\text{PV module size}} = \frac{{\text{Total Watt peak rating}}}{{\text{PV output power rating}}}. $$

(31)

Inverter sizing

The size of an inverter used for any solar photovoltaic project is a function of the total wattage of the energy consumed and the factor of safety, and it is calculated as shown below:

$$ {\text{Inverter size }} = {\text{Peak energy requirement}} \times {\text{Factor of safety}}. $$

(32)

A 100 MW photovoltaic solar system is proposed for the project to power the selected location. The Canadian Solar mono-Si-CS6X-300 M, details of which are given in Table 5, was considered because of its availability and cost-effectiveness for the Lebanese market when compared to other solar PV modules. Thus, to build the 100 MW grid-connected PV plant in the suitable region, 333,333 modules are required. The Canadian Solar mono-Si-CS6X-300 M, details of which are given in Table 4, was considered because of its availability and cost-effectiveness in the Lebanese market when compared to other solar PV modules. Moreover, the proposed inverter for the 100 MW solar power plant is a GE Power Conversion: Brilliance inverter (the specifications of the used inverter are shown in Table 5).

Table 4 Technical details of the Canadian Solar mono-Si-CS6X-300 M

Table 5 Technical details of the GE power conversion: brilliance inverter

Economic analysis

RETScreen Expert software is used in this study to evaluate the technical, economic and environmental aspects of the proposed wind/PV projects and to compare its potential between selected regions. Several researchers have worked on designing a wind/PV system using RETScreen to estimate the potential and viability of renewable projects, especially those involving wind and solar (Owolabi et al. 2019; Rafique et al. 2018; Kassem et al. 2020). RETScreen Expert software is also capable of estimating annual and monthly energy productions, capacity factors, and other important economic measures, which are defined as an equation given below:

Net present value (NPV)

$$ {\text{NPV}} = \mathop \sum \limits_{n = 0}^{N} \frac{{C_{n} }}{{\left( {1 + r} \right)^{n} }} . $$

(33)

Levelized cost of energy (LCOE)

$$ {\text{LCOE}} = \frac{{\text{sum of cost over lifetime}}}{{\text{s of electricity generated over the lifetime }}} . $$

(34)

The internal rate of return (IRR)

$$ O = \mathop \sum \limits_{n = 0}^{N} \frac{{C_{n} }}{{\left( {1 + {\text{IRR}}} \right)^{n} }}. $$

(35)

Simple payback (SP)

$$ {\text{SP}} = \frac{{C - {\text{IG}}}}{{\left( {C_{{{\text{ener}}}} + C_{{{\text{capa}}}} + C_{{{\text{RE}}}} + C_{{{\text{GHG}}}} } \right) - \left( {C_{o\& M} + C_{{{\text{fuel}}}} } \right)}} . $$

(36)

Equity payback (EP)

$$ {\text{EP}} = \mathop \sum \limits_{n = 0}^{N} C_{n} . $$

(37)

Annual life cycle savings (ALCS)

$$ {\text{ALCS}} = \frac{{{\text{NPV}}}}{{\frac{1}{r}\left( {1 - \frac{1}{{\left( {1 + r} \right)^{N} }}} \right)}}. $$

(38)

GHG emission reduction cost (GHGERC)

$$ {\text{GHGERC}} = \frac{{{\text{ALCS}}}}{{\Delta_{{{\text{GHG}}}} }} . $$

(39)

Capacity factor (CF)

$$ {\text{CF}} = \frac{{P_{{{\text{out}}}} }}{P \times 8760} . $$

(40)

Results and discussion

In this study, the characteristics of wind speed and energy potential in Rayak region, Lebanon were investigated using 11 years of wind speed measured at a height of 10 m. In addition, the wind power and energy densities were calculated at various heights to classify the wind power at the selected location. In addition, ANN models were used to identify the most significant meteorological parameters that affect the wind speed measurement. Moreover, a feasibility study of a 100 MW grid-connected wind/solar project was conducted using RETScreen Expert software. The following section presents a discussion of the results obtained.

Wind energy potential and identifying the most significant input parameters for predicting wind speed

Wind speed characteristics and wind energy potential

Figure 6 shows the seasonal daily variation of wind speed for the 11 years (1996–2006) at the selected location. It can be seen that the maximum and minimum values of wind speed during the investigation period are 9.089 m/s and 1.372 m/s, respectively. In addition, it is observed that the higher wind speeds occur in the autumn season and the minimum wind speed occurs in winter season. Moreover, the annual daily variation in wind speed during the investigation period is shown in Fig. 7. It is found that the minimum and maximum wind speeds range between 7.047 m/s and 4.201 m/s with an average value of 5.884 m/s. It is found that the highest and lowest wind speeds were recorded in 1996 and 2001, respectively. The mean annual wind speed for 1996–2006 is 5.884 m/s; therefore, it is concluded that the selected region has high wind resource potential.

Fig. 6

Seasonal daily wind speed variations for each season

Fig. 7

Annual daily wind speed variation

Furthermore, Fig. 8 represents the annual season wind speeds for the 11 years along with their average for the specific location. The whole year is divided into four seasons: winter (December to February), spring (March to May), summer (from May to August) and autumn (September to November). It is found that the maximum wind speed of 6.953 m/s occurred in winter and the minimum value was also recorded in winter with a value of 4.334 m/s. In addition, it is observed that the highest mean wind speed of 6.284 m/s was recorded in autumn followed by summer and spring. Winter has the lowest mean wind speed with a value of 5.417 m/s.

Fig. 8

Annual season wind speed for year (1996–2006)

To study the characteristics of the wind speed data, the Weibull probability distribution function was used. The wind speed frequency distribution, mean wind speed, and Weibull parameters have been calculated for the selected region. In addition, the most probable wind speed and the wind speed that carries the most energy were estimated by utilizing the parameters of the Weibull distribution function. Table 6 summarizes the seasonal and annual mean wind speed, standard deviation (SD), Weibull parameters and specific wind characteristics of the selected location for the period of 1996–2006. It is observed that the maximum and minimum values of shape parameter (k) are 22.853 in summer 2000 and 2.103 in autumn 1998, respectively. The value of k is above 2, which indicates that the wind speed was moderate to steady at a height of 10 m at the selected location. In addition, the k value for the 11 years was found to be 23.956. Moreover, the maximum scale parameter (c) was found to be 8.238 m/s in autumn 1997 and 3.009 in summer 2003. Furthermore, it is observed that the $V_{{{\text{mp}}}}$ and $V_{{{\text{maxE}}}}$ values ranged from 7.27 to 3.67 m/s and 10.83 to 3.10 m/s, respectively. In addition, the maximum value of $V_{{{\text{mp}}}}$ and $V_{{{\text{maxE}}}}$ was estimated at 7.27 m/s in winter 2006 and 10.83 m/s in autumn 1998. The $V_{{{\text{mp}}}}$ and $V_{{{\text{maxE}}}}$ for the 11 years were 5.98 m/s and 6.01 m/s, respectively, which is almost the same. The probability density functions and cumulative density functions with real data histograms are presented in Fig. 9 for 1996–2006.

Table 6 Season, annual mean wind, standard deviation (SD), Weibull parameters, and specific wind characteristics for 1996–2006

Fig. 9

Wind speed frequency distribution analysis for 1996–2006

Moreover, it is important to discuss the output generated by the wind in the form of power density. The seasonal air density at various heights was calculated using Eq. (16) and ranged from 1.171 to 1.245 kg/m³. In addition, the average annual air density of the selected location was found to be 1.221 kg/m³ at a height of 10 m. A similar result was found by Olaofe and Folly (2013). They found that the average air density of Darling City was 1.222 kg/m³ at a height of 10 m. After calculating the density at a height of 10 m, Eq. (10) was used to calculate the wind power density at that height and the results are listed in Table 7. It is found that the maximum and minimum actual wind power density values occurred in the summers of 2000 and 2003 with values of 217.389 W/m² and 13.997 W/m², respectively. In addition, it is noticed that the average annual wind speed for the period 1996–2006 was found to be 124.534 W/m². The wind energy potential classification according to the value of wind power density is presented in Table 8. From these results, it observed that the selected region could be considered as class 2, which indicates marginal wind power potential. In addition, according to Fazelpour et al. (2017), the wind density of the selected location is classified as fairly good, as shown in Table 9.

Table 7 Season and annual wind power density

Table 8 Wind power classification at the 10 m height (Ilinca et al. 2003)

Table 9 Wind power classification at the 10 m height (Fazelpour et al. 2017)

Figure 10 illustrates the seasonal wind speed and wind power density for 1996–2006. According to the wind power classification, the winter season is in class 1 as the actual and calculated wind power density is less than 100 W/m², whereas the spring and summer seasons are in class 2. Finally, the autumn season is in class 3 as the calculated power density is greater than 150 W/m².

Fig. 10

Seasonal wind speed and wind power density for 1996–2006

Most influential variables for predicting wind speed

Forecasting of wind speed is essential for designing a wind project, and projected potential future wind system feasibility. Therefore, several meteorological variables are used to determine the most effective variables for wind speed prediction. In the present study, 63 ANN models with different input combinations are evaluated for the prediction of WS. All the models are validated against the actual data of the WS based on statistical measures such as RMSE and MSE. For higher modeling accuracy, RMSE indices should be closer to zero. MATLAB 2015a was used to train and test the developed ANN models. To select the best architecture of the developed ANN model, the numbers of hidden layers and neurons were varied. The network with the minimum MSE invalidation is called the trained ANN model. The best performance of the network was obtained by training the developed ANN architecture several times until the MSE showed the minimum value. The same-trained network was tested with the new datasets to check the performance of the network. The numbers of neurons started from 1 and were constantly increased and re-training of the ANN continued until the satisfactory then is considered as the optimal number. Table 10 shows the optimum number of hidden layers and neurons and the activation function that was chosen for each ANN model. 63 ANN models (ANN1 to ANN36) were developed with different input combination and the target for all the models is daily WS.

Table 10 Evaluation of the networks and Statistical tools’ performance of the ANN model with a single input

Case 1: parameter selection for 1 input (ANN1–ANN7)

It is found that ND gave the best performance followed by Tmin for estimating the WS. Moreover, it is observed that the lowest performance was obtained from QNH with RMSE of 0.2221.

Case 2: parameter selection for 2 input (ANN8–ANN28)

21 possible combinations were considered and analyzed to find the best integration of 2 parameters. It is found that the lowest RMSE was recorded for the combinations of ANN 27 [QNH, ND] and the highest RMSE was obtained from the combinations of ANN 24 [Hmax, ND] for estimation of the daily WS.

Case 3: parameter selection for 3 input (ANN29–ANN43)

In this case, three parameters were considered as inputs. For this purpose, 15 ANN models were developed to evaluate the importance of each input variable and identify the most relevant input. The ANN 43 with a combination of [ND, QFE, QNH] produced the lowest RMSE and ANN 34 with a combination of [Tmax, Hmin, Hmax] gave the highest RMSE.

Case 4: parameter selection for 4 input (ANN44–ANN53)

In this case, the test was performed to select the most important sets combined of 4 input parameters. To achieve this, 10 different combinations of 4 inputs were considered to determine and introduce the most significant one for the prediction of daily WS. It is observed that the best combination for predicting daily WS was ANN 51 [Hmin, Hmax, Nd, QFE] with RMSE equal to 0.1583. In addition, it is found that the maximum RMSE was recorded for ANN 50 with a combination of [Hmin, Hmax, ND, QNH] for estimating daily WS.

Case 5: parameter selection for 5 input (ANN54–ANN59)

Six ANN models with five different combinations of inputs were developed to predict the daily WS. The ANN 54 tested with the combination of [Tmin, Tmax, Hmin, Hmax, ND] gave the lowest RMSE. In addition, it is noticed that the ANN 59 with the combination of [Hmin, Hmax, ND, QFE, QNH] produced the maximum RMSE value.

Case 6: parameter selection for 6 input (ANN60–ANN62)

In this case, the test was performed to select the most important sets combined of 6 input parameters. To achieve this, 3 different combinations of 6 inputs were considered to determine and introduce the most significant one for the prediction of daily WS. It is observed that the best combination for predicting daily WS was ANN 60 and ANN 61 with combinations of [Tmin, Tmax, Hmin, Hmax, ND, QFE] and [Tmain, Tmax, Hmin, Hmax, ND, QNH], respectively, with a value of 0.1235 for RMSE. In addition, it is found that the maximum RMSE was recorded for ANN 62 with a combination of [Tmax, Hmin, Hmax, ND, QFE, QNH] for estimating daily WS.

Case 7: parameter selection for 7 input (ANN63)

All seven input variables were applied to train and test the developed ANN, which was named ANN 63. The ANN-35 showed good prediction accuracy with an RMSE value of 0.1445.

Predicted daily WS values using the best input combination of all the ANN models are compared with the actual values, as shown in Fig. 11. The best combination of each ANN model is presented in Table 11. Out of the 63 ANN models, ANN 60, ANN61 and ANN 5 gave the best predictions with the combinations of [Tmin, Tmax, Hmin,Hmax, ND, QFE], [Hmin, Hmax, ND, QFE, QNH] and [ND], respectively. From the developed ANN models, the most relevant input variables for predicting the daily wind speed are found to be temperature in terms of maximum temperature and minimum temperature, pressure and relative humidity. In fact, measuring wind speed is necessary to calculate the annual energy production of a wind farm. Therefore, in the literature, several studies demonstrated that the air density influenced the power from a wind turbine (Danook et al. 2019; Yue et al. 2017; Guerrero-Villar et al. 2019; Ulazia et al. 2019). For example, Danook et al. (2019) found that the humid air implies lower density resulting in lower power from a wind turbine. Yue et al. (2017) concluded that at high temperatures, the high humidity effect on air density cannot be ignored for annual energy production calculations.

Fig. 11

Comparison between actual and predicting values by ANN 5 and ANN 60

Table 11 Rank of each ANN model

Solar energy potential

In general, two representative parameters including global horizontal irradiation (GHI) and direct normal irradiation (DNI) are used to analyze the global solar resources in the country. GHI and DNI are considered important parameters for assessing energy generation for PV/flat-plate photovoltaic technologies and the development of CSP (concentrating solar power) and CPV (concentrating photovoltaic) systems, respectively. As mentioned before, average monthly NASA data (2007–2016) were used to analyze the potential of solar energy in the selected region. Figure 12 illustrates the monthly solar irradiation in terms of GHI and DNI and the monthly mean air temperature values (AT) for the selected region. Based on the results, the monthly values of GHI ranged from 74.906 to 92.622 kWh/m². The maximum and minimum monthly values of GHI are recorded in July and December, respectively. Furthermore, it is found that the highest monthly DNI value of 249.565 kWh/m² was obtained in July. In addition, it is noticed that the maximum and minimum air temperature values were recorded in July and January with values of 23.28 °C and 8.55 °C for the selected region. Furthermore, the annual GHI and DNI were estimated to be 1877.41 kWh/m² and 1967.98 kWh/m², respectively. According to Prăvălie et al. (2019), it is found that the selected region has high solar resources and is categorized as having excellent potential. Consequently, it is concluded that the Rayak region is a suitable location for installing a large-scale photovoltaic system.

Fig. 12

Mean monthly GHI, DNI and AT for the selected region

Comparative techno-economic assessment and environmental impacts of wind/solar projects

RETScreen software is used to investigate the feasibility of a 100 MW installed capacity wind farm and solar plant project in the selected region. In this section, a comparison between the two projects is made. Figure 13 presents the monthly results for the electricity generation from the installed projects. It is found that the highest electricity generation of 349717 MW is produced by the wind farm compared to the solar plant (i.e., 1478.053 MW). In addition, it is observed that the maximum electricity production was recorded in December for the wind farm and June for the solar farm. Moreover, the capacity factor of each project was 39.9% for the wind farm and 16.1% for the solar plant project.

Fig. 13

Monthly power exported to grid for wind and solar projects

The financial parameters of the proposed system seem to be economically feasible, as listed in Table 12. An overview of the cash flow over the project life time for the wind farm and solar plant projects for the selected region is shown in Fig. 14. The cash flow indicates that investors are expected to receive a positive cash flow from year 8.6 onwards for the wind farm project and year 21.3 onwards for the solar plant project. A minimum equity payback of 6 years is observed for the wind farm project and 13.5 years for the solar plant project.

Table 12 Financial parameters

Fig. 14

Cumulative cash flow for all projects

Moreover, the economic performance of the 100 MW installed capacity wind/solar project for the selected location is presented in Table 13. The results show that the proposed 100 MW wind farm project is very promising in the selected location due to the economic performance results obtained. Furthermore, the wind farm is a more economical option than the solar plant because of the higher values of NPV, BCR, ALCS, and IRR as well as the lower values of EB, SB and LCOE.

Greenhouse gas (GHG) emissions of the proposed renewable projects in terms of the total amount of annual GHG reduction in terms of tons of CO₂ per year (tCO₂/year) and cost ($/tCO₂) were estimated using RETScreen software. The utilization of wind energy as a partial replacement for traditional power production units in the country significantly reduces the emissions of GHG, as can be observed from Table 14. The obtained results demonstrate that a significant amount of CO₂ could be reduced by the proposed projects in the studied location. It can be observed that the total amount of CO₂ that can be reduced from the proposed projects was 245,472 tCO₂/year from the wind farm and 84,910,458 tCO₂/year from the PV plant. Generally, it can be concluded that the wind farm project reduce increased amounts of CO₂ emissions compared to the PV plant project.

Table 13 Performance of 100 MW installed capacity of wind farm and solar plant projects

Table 14 Net annual reduction of GHG emissions

Conclusion

Both population and industrial growth are causing significant increases in energy demand throughout over Lebanon, especially in terms of electricity demand, which means there is a considerable need to find alternative sources to replace the traditional energy sources produced by fossil fuels. Therefore, the present study was focused on wind and solar power potential assessment for the Rayak region in Lebanon.

For wind energy potential, the two-parameter Weibull distribution function is used to represent the wind speed distribution for the Rayak region in Lebanon. Wind speed data were collected from a measurement station in the Rayak area for an 11-year period measured at a height of 10 m. Yearly and seasonal parameters of the Weibull distribution function were estimated using the daily average wind speed data with the Maximum likelihood method. In addition, wind power density was calculated. For this purpose, the air density values were estimated using collected pressure and temperature values. As a result, the average annual wind speed for the years 1996–2006 was found to be 124.534 W/m², which could be considered as class 2 and indicates marginal wind power potential. Furthermore, the objective of the study was to predict the daily average wind speed using an Artificial Neural Network (ANN) approach. In this study, 63 ANN models were developed by varying the weather parameters. An 11-year database comprising daily wind speed, minimum temperature, maximum temperature, minimum relative humidity, maximum relative humidity, altimeter Pressure Settings and atmospheric pressure at aerodrome elevation were used in the ANN models. All the models were validated and the performances of the models were analyzed. Out of the 63 ANN models, ANN 60, ANN61 and ANN 5 gave the best prediction with the combinations of [Tmin, Tmax, Hmin, Hmax, ND, QFE], [Hmin, Hmax, ND, QFE, QNH] and [ND], respectively. Besides, for the future installation of PV, the solar energy potential of the selected location was assessed using NASA's average monthly solar radiation. The results indicated that Rayak region has high solar resources and are categorized as excellent. Consequently, it is concluded that Rayak region is a suitable location for installing large-scale photovoltaic systems.

Moreover, the purpose of the present study is to evaluate the feasibility of developing 100 MW grid-connected wind/PV power project in the selected region. To fulfil this objective, RETScreen Expert software was used to validate the techno-economic and environmental sustainability of installing a grid-connected wind and PV system in the country. mono-Si-CS6X-300 M module manufactured by Canadian Solar and AN BONUS 1 MW-70 m wind turbine manufactured by Siemens Company were selected as an efficient PV module and wind turbine for the proposed solar PV plant and wind farm project, respectively. The results showed that the large-scale wind farm is a more economical option than the solar PV plant because of the higher values of NPV and ALCS as well as the lower values of EP and LCOE.