International Journal of Applied Power Engineering (IJAPE)

Vol. 13, No. 1, March 2024, pp. 91~101

ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i1.pp91-101  91

Techno-economic assessment and wind energy potential of

Nagad in Djibouti

Abdoulkader Ibrahim Idriss1, Ramadan Ali Ahmed2, Hamda Abdi Atteyeh1,

Abdou Idris Omar1, Tahir Cetin Akinci3
1
Department of Electrical and Energy Engineering, Faculty of Engineering, Université de Djibouti, Djibouti, Djibouti
2
Department of Electrical and Energy Engineering, Institute of Industrial Technology, Université de Djibouti, Djibouti, Djibouti
3
Department of Electrical Engineering, Istanbul Technical University, Istanbul, Turkey

Article Info ABSTRACT

Article history: The use of small scaled horizontal and vertical axis wind turbines in urban
installation is increasing over the world. However, in Djibouti, the latter is
Received Jun 12, 2023 still in the development phase. The paper presents a techno-economical
Revised Jul 27, 2023 analysis and wind energy potential for the period of five years (2015-2019)
Accepted Aug 2, 2023 in Nagad based on actual measured wind speed data collected every 10 min
at 10 m height. The energy pattern factor method has been used to estimate
the Weibull parameters. With this method, the mathematical complexity is
Keywords: reduced with a minimization of the error at any heights and locations when
calculating the wind power density. At 50 m height, the shape parameter
Economic analysis showed a small variation for different periods. The scale parameter values of
Nagad 7.78 m/s and 4.8 m/s were obtained in the hot and cold seasons, respectively.
Wind power The results showed that the Nagad site is suitable for wind power
Wind speed development. According to the economic viability, RX30, Vestas V20,
Wind turbine Enercon, Nordex N27, and Vestas V44 wind turbines are recommended for
the Nagad site due to their low energy price ranging from 0.05$/kWh to
0.31$/kWh. This is 2-6 times cheaper than the average local tariff of
electricity in Djibouti.
This is an open access article under the CC BY-SA license.

Corresponding Author:
Abdoulkader Ibrahim Idriss
Department of Electrical and Energy Engineering, Faculty of Engineering, Université de Djibouti
Street Djanaleh, BP 1904, Djibouti, Djibouti
Email: [email protected]

1. INTRODUCTION
The future of the East African population is uncertain due to two major challenges which are the
lack of access to modern energy services and the vulnerability associated with climate change. These
challenges are at the origin of the reflections engaged to reinvent the energy future by carrying out the
necessary transitions towards energy systems that allow responsible growth and reconcile economic
development, environmental protection, and the reduction of inequalities.
Access to electricity is a key indicator of a country's level of development. Generation of electricity
from renewable energy such as wind, sun (solar thermal and photovoltaic), hydro, and geothermal can play a
major role in electricity production in Eastern African countries. Several programs are defined by the
Sustainable Energy for All [1], [2] as well as the renewable capacity statistics from the International
Renewable Energy Agency [3] to offer a new form of planning centered on needs and to redefine the energy
model of the region and the associated policies. Wind energy is the most non-polluting, sustainable and can
potentially make a significant contribution to developing countries with poor infrastructure for power
generation. The cost, performance, and reliability of renewable energy technologies are significantly

Journal homepage: http://ijape.iaescore.com

92  ISSN: 2252-8792

improved to the point that they can now compete with conventional energy sources in several applications
[4], [5]. Numerous studies have been done to assess the wind speed characteristic and wind power potential
in the world [6], [7] and especially in Africa [8]–[10]. According to [11], the levelized cost of electricity and
net present cost in Yanbu region of Saudi Arabia are estimated as (0.0885$/kWh and 23.8$) for Enercon
E-126 EP4 wind turbine that leads their corresponding values of (0.142$/kWh and38.3$) for WES 30 turbine.
Abd in [12] has provided a strategy based on a weather change to find the optimal designing and modelling
for four types of wind energy conversion system models using HOMER software. The study has focused on
the technical, economic, and profitability calculation for any renewable energy system. In Saswat et al. [13],
India have demonstrated the effectiveness of the hybrid PV/solar/wind power system, which is given the best
and most efficient alternative to conventional energy sources. Idriss et al. [14] conducted the potential of
wind and solar energy in two rural sites in Djibouti which are Herkalou and Lake Assal. They showed that
the studied sites receive the greatest amount of solar radiation compared to other places in the world with the
global radiation value of 2,898 kWh/ (m². year). In addition to that, the sites have an encouraging potential to
develop a hybrid power system for any application. Queen et al. [15] have investigated the advantages of
integrating natural source of energy from the renewable energies to the prevailing electric power systems. For
the two studied standard IEEE system (IEEE 14 bus and IEEE 30 bus), the price of electricity acquired from
the grid is lowered by 30% with incorporated renewable energy systems. In addition, Sakhrieh et al. [16]
provided a techno-economical study of the optimized hybrid system includes photovoltaics, a biogas
generator, batteries, and a diesel generator in the rural sites. The levelized cost of energy of 0.06$/kWh and a
net present cost of 2,100,000$ have been estimated for the optimized hybrid system. In Egypt,
Abdelrahman et al. [17] have examined the wind energy potential and the economic feasibility to develop the
first wind farm at Elkharga Oasis. With 50 MW wind farm at Elkharga, they concluded that the cost of
energy is much cheaper by > 50% than the current tariff in Egypt. Daoudi et al. [18] have demonstrated the
economic viability of the two onshore wind farms located in the province of Tantan. The results have shown
that the two wind farms have a good potential to develop the wind farm with the cost of the production values
of 3.45$/kWh and 3.87$/kWh in Tantan-1 and Tantan-2, respectively.
The total potential of wind energy is estimated at 1,300 GW in the Sub-Saharan African regions.
The use of micro and small wind turbines in urban and rural installations is increasing over the world, and it
is still in the developing phase in East Africa. Djibouti imports all its energy needs while the country has a
high potential in renewable energies, untapped until now. Recent surveys conducted as part of the strategy to
combat poverty reveals that 49.7% of sedentary households (99.5% of which are in urban area) use electricity
to power lighting systems with an average consumption estimated at 228 kWh/year per capita. However, this
prominence of energy for Djiboutian households’ masks huge disparities in access linked to the availability
and high production costs of around 52 DJF/kWh (0.32$/kWh set by Electricité de Djibouti). The country
survived until May 2011 on electricity produced from imported petroleum products. Consequently, high
production costs are a barrier to access to energy sources for the poorest of the population. In the Republic of
Djibouti, a few studies have been conducted to analyze wind power generation [19], more attention should
concentrate on the resource of wind speed potential in the urban, peri-urban, and rural areas of the country.
During the implementation of the wind project, it is recommended to evaluate the character of the wind speed
data, the feasibility depending on the location, the characteristics of wind turbines (horizontal or vertical axis
models, cut-in and cut-out velocity, rated velocity, power energy output, and capacity factor), the cost
analysis (initial cost, maintenance costs during the lifetime of the turbine), and the energy potential before
any wind energy system. Researchers have studied the integration of micro, small, and mid-sized wind
systems using several statistical and probability distribution analyses of wind speed data [20]–[22].
According to the International Electrotechnical Commission (IEC:614-00-12), the Weibull 2-parameters are
becoming a standard indicator of probability distribution function (PDF) to describe the wind characteristics
[23]. Others have investigated by using several numerical methods for fitting the wind speed data as well as
the graphical (GM), the moment (MM), and the energy pattern factor (EPFM) methods [24]–[26].
In this work, the wind potential assessment, and the economic feasibility of using commercially
available wind turbines are evaluated for the Nagad peri-urban site, located in the southern part of Djibouti-
city. Potential Djibouti sites for wind energy generation have not been thoroughly explored as the
development of wind project continues to be hampered by the lack of reliable and accurate wind datasets in
many parts of Djibouti, as well as the lack of both qualified human resources and accessibility of
mountainous and hostile areas of the northern and southern regions for scientists and researchers. Therefore,
the objective of this paper is to evaluate the cost of energy production from micro, small and mid-sized wind
turbines which will serve as a benchmark in the national energy plan. The statistics and costs provided by this
study have enabled government officials and potential investors to propose a strategy to reduce the weight of
energy consumption bills in the household budget and to make energy more accessible for all. The
fundamental contributions and research originality of this paper can be summarized as:

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101
Int J Appl Power Eng ISSN: 2252-8792  93

- Estimating the wind energy, for the first time, in Nagad owing to its windy and less strict topography
locations
- Illustrating the performance comparisons of nine wind turbines with various technologies (horizontal and
vertical) and characteristics
- Evaluating the economic viability by analysing the cost of energy production
The rest of this paper is subdivided into four sections. Section 2 presents the methodology including
the site description and the wind data analysis are presented. In section 3, the results are discussed. Finally,
the section 4 presents the conclusion and the recommendation of this study.

2. METHOD
2.1. Study area: Nagad site description and wind data
Nagad is a coastal site (at 11.3124° N and latitude 43.0739° E, altitude 8 m) located near to the
International Airport of Djibouti. The site is situated in the south of Djibouti city and is identified as a peri-
urban area. Due it’s to rather close proximity to the equator, Djibouti is classified as a hot and humid country.
This type of climate receives the highest amount of solar radiation compared to other regions. The high level
of solar radiation also causes high air temperatures. The meteorological actual wind data were collected and
analyzed hourly and every 10 min taken by a mast at 10 m height for a period of five years from 2015 to
2019 using Vantage Pro2 equipment. It includes a mast, an anemometer, a wind vane, a thermometer, and a
barometer. The equipment was installed in July 2014 on the roof of the Department of Electrical and Energy
in the Faculty of Engineering. Daily average wind speed, wind directions, and temperatures were measured in
10 min time intervals, at 10 m height above the ground level.
Figure 1(a) displays a histogram that illustrates the monthly average temperature (including
minimum and maximum). This graph highlights the cold and hot seasons. The period extends from October
to April where the climate is rather pleasant, with average temperatures between 18 °C and 34 °C during the
day and 17 °C and 32 °C at night. The second season extends from May to September, with very high
temperatures. Ranging from 24 °C to 46 °C during the day, and at night it ranges from 20 °C to 42 °C. The
difference between the minimum and maximum temperatures in the hot summer months is slightly higher
than in the cooler months.
Figure 1(b) shows a monthly average contour map of diurnal mean wind speeds over the studied
period. The average wind speeds are higher during the hot period (5-7.5 m/s from 12 p.m. to 8 p.m.) than
during the coldest one (4-6 m/s from 12 p.m. to 8 p.m.). The wind characteristics information for a specific
site can be combined visually in a wind rose. The latter allows the three essential pieces of information to be
grouped in a single graph: wind speed, direction, and frequency by sector.
Figure 1(c) reveals an omnidirectional wind with a dominant wind from the west (5%, 9 m/s)
generated by the Ethiopian highlands in summer. This wind which is called "Khamsin" is dry and hot. During
the cold season, the city is swept by easterly winds (11.5%, 11 m/s) generated by the trade winds, humid
winds from Arabia, and the Gulf of Aden. The annual mean wind speed is 4.28 m/s during the period
considered, here 5 years. These representations are the first approach, allowing to have a quick overview of
the wind profile of a site for a wind energy application.
Further analyses of wind data were performed in Figure 1(d). The EPFM method is adopted in this
study to describe the power density function (PDF) and the cumulative density function (CDF). The EPFM
method requires less computation, and easier implementation to calculate the wind power density with less
error at any height and location. Based on the latter method, the scale and shape parameters are estimated,
which will be discussed in section 3.

2.2. Wind data analysis

2.2.1. Weibull parameters and energy pattern factor method (EPFM)
Several mathematical models have been used to assess the wind speed. In this investigation, the
2-parameters Weibull function is chosen, because it gives a good fit and better measurement of probability
distribution function than other statistical methods [27]. The PDF and the CDF of the 2-parameters Weibull
can be expressed as:

𝑘 𝑣 𝑘−1 𝑣 𝑘
𝑓(𝑣) = ( ) ( ) 𝑒𝑥𝑝 [− ( ) ] (1)
𝑐 𝑐 𝑐

𝑘 𝑣 𝑘−1 𝑣 𝑘
𝑓(𝑣) = ( ) ( ) 𝑒𝑥𝑝 [− ( ) ] (2)
𝑐 𝑐 𝑐

where v is the wind speed, c (scale) and k (shape) are the Weibull parameters, respectively.
Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss)
94  ISSN: 2252-8792

To calculate k and c values, several methods are proposed in the literature [28], [29]. In this work,
the EPFM is selected because it is a valuable and direct method that does not demand repetitions, simple and
̅̅̅3 ) to the cube
easy to tool and formulate. 𝐸𝑝𝑓 is defined the as ratio between the mean of cubic wind speed (𝑣
3
of mean wind speed (𝑣 ). The 𝐸𝑝𝑓 can be calculated as (3) [21].
1 𝑛
̅̅̅̅
𝑣3 ∑ 𝑣3
𝑛 𝑖=1 𝑖
𝐸𝑝𝑓 = 3 = 1 3 (3)
𝑣 ( ∑𝑛 𝑣)
𝑛 𝑖=1 𝑖

Then, the values of k and c are determined by (4) and (5).

3.69
𝑘 =1+ 2 (4)
(𝐸𝑝𝑓 )

𝑣
𝑐= 1 (5)
𝛤 (1+ )
𝑘

With Γ(.) is the gamma function.

(a) (b)

(c) (d)

Figure 1. Nagad site wind data covering 2015-2019 (a) monthly mean temperature of Nagad’s wind speed
data, (b) monthly (and daily diurnal) contour map of the hourly mean wind speed data, (c) wind rose
diagram, and (d) PDF and CDF curves compared to the observed wind data

2.2.2. Wind speed extrapolation with height

The wind speed increases with heights, the power law is used for its simplicity in this study to
extrapolate wind speed at different altitudes [30]. The wind speed variation with height can be
mathematically expressed as (6).
ℎ 𝛼
𝑣(ℎ) = 𝑣0 ( ) (6)
ℎ0

Where 𝑣0 is the wind speed at the initial height ℎ0 and α is the power-law commonly admitted to be 1/7. As
the wind speed varies with height, similarly, the Weibull parameters c and k are also functioning of hub
height. Using (7) and (8), the parameters can be calculated.

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101
Int J Appl Power Eng ISSN: 2252-8792  95

ℎ 𝑛
𝑐(ℎ) = 𝑐0 × ( ) (7)
ℎ0

ℎ
(1−0.088𝑙𝑛( ))
ℎ0
𝑘(ℎ) = 𝑘0 × ℎ (8)
(1−0.088 𝑙𝑛( ))
10

Where 𝑐0 and 𝑘0 are Weibull parameters at ℎ0 . The exponent 𝑛 is given by (9).

(0.37−0.088𝑙𝑛(𝑐0 ))
𝑛= ℎ (9)
(1−0.088 𝑙𝑛( ))
10

2.2.3. Evaluation of wind power density, energy density, capacity factor, power, and accumulated
annual energy of the wind turbines
The wind power is mathematically expressed as (10) [31].
1 3
𝑃(𝑣) = 𝜌𝐴 𝑣 (W) (10)
2

The air density ρ is assumed to be 1.225 kg/m3 and the rotor area of the turbine is A (m²). The wind power
density (WPD) in a selected site at a period with Weibull parameters can be expressed as (11).

𝑃 (𝑣) 1 3
𝑊𝑃𝐷 = = 𝜌 𝑐 3 𝛤 (1 + ) (𝑊/𝑚²) (11)
𝐴 2 𝑘

While wind energy density (WED) is defined as the power density over a period, energy density can be
evaluated using (12). Where T is a period of time (hour).

𝑊𝐸𝐷 = 𝑊𝑃𝐷 × 𝑇 (𝑊ℎ/𝑚²) (12)

To evaluate the efficient and best-suited wind turbine for the Nagad site, nine commercial turbines
of the site were selected. Average power output (𝑃𝑒.𝑎𝑣𝑒 ), capacity factor (CF), and accumulated annual energy
(AEP) are the important performance parameters to study wind speed installed in a given site. The 𝑃𝑒.𝑎𝑣𝑒 and
CF of a wind turbine are calculated by (13) and (14).

𝑣 𝑘 𝑣 𝑘 𝑣𝑓 𝑘
−( 𝑐𝑐 ) −( 𝑟 )
𝑒 −𝑒 𝑐 −( )
𝑃𝑒.𝑎𝑣𝑒 = 𝑃𝑒𝑅 ( 𝑣 𝑘 𝑣 𝑘
−𝑒 𝑐 ) (13)
( 𝑟) −( 𝑐 )
𝑐 𝑐

𝐶𝐹 = 𝑃𝑒.𝑎𝑣𝑒 /𝑃𝑒𝑅 (14)

Where 𝑣𝑐 , 𝑣𝑟 and 𝑣𝑓 are the cut-in, rated, and cut-off wind speeds respectively. PeR is the rated electrical
power of the wind turbine. The AEP is calculated over a period by using (15) [32]. Where t is the time,
example for one a year period it has 8760 in hours.

𝐴𝐸𝑃 = 𝐶𝐹 × 𝑃𝑒𝑅 × 𝑡 (𝑘𝑊ℎ) (15)

2.2.4. Wind energy cost analysis

In a given site, to assess the feasibility of the wind farm, the cost of energy is the most important
parameter to evaluate the economic viability. For evaluating wind energy cost, several methods have been
discussed in [33]. However, the present value of costs (PVC) method is commonly used, and it is adopted in
this study. The PVC is given as (16).

1+𝑖 1+𝑖 𝑙𝑡 1+𝑖 𝑙𝑡

𝑃𝑉𝐶 = 𝐼 + 𝐶𝑜𝑚𝑟 [ ] × [1 − ( ) ]−𝑆( ) ($) (16)
𝑟−𝑖 1+𝑟 1+𝑟

The cost of produced energy (in kWh) by turbines at the respective location was estimated as 20%
for the investment cost (I), the inflation (i) and interest (r) rates are 2% and 11.2% [34], with the lifetime (lt)
of wind turbines is 20 years. The operation, maintenance, and repair cost (Comr) are 15% (minimum cost)
and 25 % (maximum cost). The scrap value (S) is 10 %. The cost per kWh of electricity generated (UCE) can
be determined by (17) [33].
Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss)
96  ISSN: 2252-8792

𝑃𝑉𝐶 $
𝑈𝐶𝐸 = ( ) (17)
𝐴𝐸𝑃 𝑘𝑊ℎ

3. RESULTS AND DISCUSSION

3.1. Wind speed analysis: analysis of Weibull parameters, power, and energy densities
To evaluate the availability of wind power at a site, the plot of the probability density function
(PDF) curve is essential. The annual average k and c Weibull parameters are calculated using the probability
distribution for the considerate wind speeds, which are then plotted with the cumulative density function
curve see Figure 1(d). The most probable wind speed occurs at a speed of 2.5 m/s with a probability of
16.2%, while wind speed greater than 10 m/s shows a very low probability. Table 1 presents the monthly,
seasonal, and yearly average of wind speed, the Weibull parameters (c and k), and the power and energy
densities (WPD and WED) for the studied site at 10 m height.
The results show that the whole year is divided into two seasons: cold (from October to April) and
hot (from May to September) seasons. The highest mean wind speeds were observed in July and August with
values of 4.99 m/s and 5.30 m/s, respectively. The lowest was obtained in April with a value of 3.38 m/s.
The EPFM value of parameter k ranges from 1.71 in April to 1.99 in August while parameter c
varies between 3.38 m/s in April to 5.30 m/s in August. The higher values of c are observed during the hot
season (5.08 m/s) and lower during the cold season (4.59 m/s). The mean seasonal values of k parameter are
observed to be 1.72 and 1.65 corresponding to the cold and hot seasons, respectively. In addition, the lower
values of wind power densities and energies are obtained as 68.32 W/m² and 49.19 kWh/m² respectively in
April while the higher values of 222.56 W/m² and 164.84 kWh/m² are computed in July, respectively. Also,
the yearly mean power and energy densities are 111.49 W/m² and 976.65 kWh/m²/year, respectively.

Table 1. Mean wind speed, Weibull parameters, WPD and WED over the considered period
Period 𝒗 (m/s) k (-) c (m/s) WPD (W/m²) WED (kWh/m²)
Jan 4.7 1.91 5.6 125.52 93.39
Feb 3.99 1.83 4.82 93.87 63.08
Mar 3.89 1.82 4.58 92.73 68.99
Apr 3.38 1.71 4.33 68.32 49.19
May 3.39 1.72 4.13 68.53 50.99
Jun 3.89 1.83 4.59 94.05 67.71
Jul 4.9 1.98 5.42 222.56 164.84
Aug 5.30 1.99 6.11 222.06 165.21
Sep 3.39 1.72 4.33 68.53 49.34
Oct 3.99 1.83 4.82 94.09 70.00
Nov 3.89 1.83 4.59 93.78 67.52
Dec 3.99 1.83 4.82 93.94 69.89
Cold season 4.10 1.72 4.59 94.88 482.78
Hot season 4.54 1.65 5.08 136.10 499.78
Annual 4.05 1.83 4.84 111.49 976.65

3.2. Monthly mean wind speed and WPD at 10 m and 50 m

The WPD classification is commonly used [35] and is established for classifying the wind
potential. The wind power resource can be divided into 7 categories from poor (class=1) to excellent
(class=5, 6, and 7) [7], [36]. According to Table 2 and as shown in Figure 2(a), for the July and August,
Nagad site shows good wind energy with WPD values higher than 600 W/m² (𝑣 =7.2 m/s) at 50 m height.
The mean WPD is higher than 350 W/m² in January while for the other months it ranges from 200 to
300 W/m² (with 𝑣 varies from 4.5 to 5.3 m/s).
In Figure 2(b) the annual wind energy density was recorded at 339 W/m² (moderate, class 3 with
𝑣 =5.4 m/s) while it was measured at 297 W/m² (marginal, class 2 with 𝑣 =5.2 m/s) in the cold season and
400 W/m² (moderate, class 3 with 𝑣 =5.7 m/s) in the hot season. From the above classification, the Nagad
site has reasonable wind energy resources and is suitable for harnessing wind turbines applications.

Table 2. Classification of WPD [35], [36]

Wind class At 10 m height At 50 m height Wind class At 10 m height At 50 m height
𝑣 (m/s) WPD (W/m²) 𝑣 (m/s) WPD (W/m²) 𝑣 (m/s) WPD (W/m²) 𝑣 (m/s) WPD (W/m²)
1 <4.4 <100 <5.6 <200 5 6.0-6.4 250-300 7.5-8.0 500-600
2 4.4-5.1 100-150 5.6-6.4 200-300 6 6.4-7.0 300-400 8.0-8.8 600-800
3 5.1-5.6 150-200 6.4-7.0 300-400 7 >7.0 >400 >8.8 >800
4 5.6-6.0 200-250 7.0-7.5 400-500

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101
Int J Appl Power Eng ISSN: 2252-8792  97

(a) (b)

Figure 2. Extrapolation for (a) monthly and (b) annual and seasonal results of mean wind speed and power
density at selected heights

3.3. Performance of nine small wind turbines: monthly, yearly and seasonal analysis
In this part of the study, the performances of the nine different wind turbines were calculated. These
turbines are selected based on their availability in the Republic of Djibouti and for their tower heights. The
characteristics of the wind turbines are given in Table 3. Here, HAWT and VAWT mean the type of
Horizontal and Vertical Axis Wind Turbines, respectively. Seven HAWTs namely Aeolos, Antaris, RX30,
Vestas V20, Enercon, Nordex, and Vestas V44, and two VAWTs namely Turby and Sun surf were used to
estimate the energy production for all the considered years and were analyzed.

Table 3. Characteristics of selected commercial wind turbines

Type Turbine model Vc (m/s) Vr (m/s) Vf (m/s) Rotor diameter (m) Hub height (m) PeR (kW)
HAWT Antaris 2.8 13 25 4 12 3.50
Aeolos 3.5 12 25 8 18 10
RX30 3 10 25 16 18 30
Vestas V20 5 17.50 25 20 24 100
Enercon 3 12 25 16.20 30 55
Nordex N27 3 13 25 27 30 150
Vestas V44 5 17 20 44 50 600
VAWT Turby 3.5 12 14 2 15 2.5
Sun surf 1.8 8 25 9 18 10

The annual and seasonal mean variation of 𝑣̅ , k, and c are shown at different hub heights
corresponding to the wind turbines heights 12 m, 15 m, 18 m, 24 m, 30 m, and 50 m. At 50 m, the mean wind
speed value is 5.37 m/s, k and c values are 1.96 and 7.41 m/s, respectively. The shape parameter shows a
small variation for different heights and periods, whereas the scale parameter values vary between 5.30 m/s
(at 12 m height) to 7.78 m/s at 50 m in the hot season. For the cold season, the c parameter variation ranges
from 4.8 m/s (at 12 m height) to 7.15 m/s at 50 m, which is a sign that the site is suitable for
wind applications. For all heights, it is noted that the hot season presents high wind characteristics than the
cold season.
Thus, Pe,ave, CF, and AEP given by these wind turbines are calculated using (13), (14), and (15).
Additionally, Figure 3 shows the monthly variation of estimated CF and AEP of wind turbines. For the
monthly analysis, the outcomes can be listed as:
- The mean capacity factors and annual energy output values were recorded in August, as 28.97% and
215 kWh for Aeolos small-scale wind turbine, whereas the minimum values were 22.80% and 59.3 kWh
for Antaris turbine, given in Figure 3(a). In this case, depending on the hub height of the wind turbine and
PeR of the turbine, the Aeolos turbine produces maximum energy compared to Antaris.
- Figure 3(b) depicts the mean CF and AEP for the VAWT turbines. The maximum mean CF value is
obtained as 50.86% in August and a minimum value of 27.29% in April for the Sun surf turbine. The
maximum mean AEP value is obtained as 378.4 kWh in August while the minimum value is computed as
196.6 kWh in April. The result shows that the Sun surf turbine generates more energy than the Turby
wind turbine.
- For the Enercon turbine, the mean AEP was maximum for August with 1515 kWh and least for April with
713 kWh as described in the Figure 3(c). The highest mean CF occurs in August with a value of 37.03%
and the lowest mean CF occurs in April with a value of 17.99%. On observing the RX30 turbine, it can be

Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss)
98  ISSN: 2252-8792

observed that the maximum AEP available in August was 899.2 kWh (with CF=40.28%) and the
minimum mean AEP value of 402.8 kWh (with CF=18.64%) was present in April.
- The mean CF and AEP generated by different mid-sized wind turbines namely Nordex, Vestas V20, and
Vestas V44 at three hub heights are estimated and shown in Figure 3(d). In August, the highest capacity
factor was 32.33% for the Nordex turbine followed by 21.22% for Vestas V44 and 13.20% for Vestas
V20. Also, for the same month, the minimum AEP values of about 3.60 MWh, 9.82 MWh, and
9.47 MWh are observed for the Nordex, Vestas V20, and Vestas V44 respectively.

(a) (b)

(c) (d)

Figure 3. Monthly variation of CF and AEP for (a) Aeolos and Antaris, (b) Sun surf and Turby,
(c) Enercon and RX30, and (d) Nordex, Vestas V44, and Vestas V20

3.4. Cost of energy production: comparison and analysis

Table 4 shows the annual and seasonal computed values of PVC ($) and the UCE ($/kWh) while the
corresponding cost of Comr are 15% and 25% of the total investment. The results show that the maximum
PVC ranged from 18136.79$ to 47953.69$ for RX30 and Vestas V44 wind turbines. The Antaris turbine
produces the highest annual cost of 2.31$/kWh and the Vestas V44 generates the lowest annual cost of
0.034$/kWh, followed by Nordex N27 with a cost of 0.060$/kWh. As indicated in Table 4, the highest cost
of electricity in the cold season was 6.28$/kWh for the Antaris wind turbine, and the lowest was 0.09$/kWh
for the Vestas V44. Hence, with regards to the hot season, the highest value of UCE was 3.42$/kWh for
Antaris, and the lowest value of UCE was 0.05$/kWh for Vestas V44.

Table 4. Results of PVC (in $) and UCE (in $/kWh) for each turbine
(min. and max. represent 15% and 25% of the total investment)
Type Turbine model PVC($) Annual Cold season Hot season
UCE ($/kWh) UCE ($/kWh) UCE ($/kWh)
Min. Max. Min. Max. Min. Max.. Min. Max.
HAWT Antaris 19691.89 20447.24 2.22 2.31 6.05 6.28 3.30 3.42
Aeolos 20849.23 21648.97 0.67 0.69 1.81 1.88 0.99 1.03
RX30 17466.80 18136.79 0.12 0.13 0.33 0.35 0.19 0.20
Vestas V20 28653.04 29752.12 0.21 0.22 0.61 0.63 0.30 0.31
Enercon 19481.20 20228.55 0.081 0.085 0.21 0.22 0.12 0.13
Nordex N27 32780.19 34037.58 0.058 0.060 0.15 0.16 0.08 0.09
Vestas V44 46182.22 47953.69 0.033 0.034 0.08 0.09 0.04 0.05
VAWT Turby 16302.32 16927.65 2.35 2.44 6.34 6.58 3.54 3.68
Sun surf 37179.34 38605.47 0.59 0.62 1.53 1.58 0.93 0.97

The local average electricity cost value is 0.32$/kWh in Djibouti. Comparing the estimated tariffs
obtained by the RX30, Vestas V20, Enercon, Nordex N27, and Vestas V44 wind turbines with the local cost

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101
Int J Appl Power Eng ISSN: 2252-8792  99

price of electricity (0.32$/kWh), indicate that the estimated tariffs are lower than the local cost price of
electricity. The latter remark is highlighted in bold in Table 4. Their cost reveals that the horizontal wind
turbines are viable renewable energy sources to serve local communities and to intensify the production of
electricity in the Nagad site.

4. CONCLUSION AND RECOMMENDATIONS

The performance comparisons and techno-economic analysis of HAWT and VAWT wind turbines
in a hot climate have been studied at Nagad peri-urban site in Djibouti city. The wind data covering
2015-2019, recorded every 10 min, were analyzed for the wind speed characteristics and economic feasibility
of wind systems. The results show that the average wind speeds are higher during the hot period (5-7.5 m/s
from 12 p.m. to 8 p.m.) than the coldest one (4-6 m/s from 12 p.m. to 8 p.m.) at 10 m hub height. The annual
mean wind speed over the considered period was 4.05 m/s. The performance of the selected empirical EPFM
method was observed and was close to the measured wind speed data of Nagad k values ranges from 1.65 to
1.99 while c values vary between 4.13 m/s to 6.11 m/s at 10 m. At 50 m, the annual mean wind speed was
5.37 m/s, k was 1.96 and c was 7.41 m/s. For the cold season, c was 7.15 m/s, which is a sign that the Nagad
site is suitable for wind power technologies development. At 50 m hub height, the annual WED was recorded
at 339 W/m² while it was measured at 297 W/m² in the cold season and 400 W/m² in the hot season. It is
important to note that Nagad site was classified as class 3 and was considered suitable to harness wind
power. In August, the highest CF was 32.33% for the Nordex turbine followed by 21.22% for Vestas V44
and 13.20 % for Vestas V20. The minimum AEP values of about 3.60 MWh, 9.82 MWh, and 9.47 MWh are
observed for the Nordex, Vestas V20, and Vestas V44, respectively.
The economic analysis showed that RX30, Vestas V20, Enercon, Nordex N27, Vestas V44 wind
turbines are recommended for Nagad due to a low price of energy between 0.05$/kWh and 0.31$/kWh. For
Vestas V44, Nordex N27, and Enercon, the price of energy is 2-6 times cheaper than the average local tariff
of electricity. The vertical turbines are not recommended for the Nagad site.

ACKNOWLEDGEMENT
The authors of this article wish to acknowledge the financial support from the University of Djibouti
and the contributions of Eng. Omar Abdoulkader Mohamed (INSTAD).

REFERENCES
[1] International Bank for Reconstruction and Development/The World Bank, “Tracking SDG7: The Energy Progress Report,” 2018.
[Online]. Available: https://openknowledge.worldbank.org/handle/10986/29812.
[2] D. L. McCollum et al., “Connecting the sustainable development goals by their energy inter-linkages,” Environmental Research
Letters, vol. 13, no. 3, p. 033006, Mar. 2018, doi: 10.1088/1748-9326/aaafe3.
[3] N. S. Ouedraogo, “Opportunities, Barriers and Issues with Renewable Energy Development in Africa: a Comprehensible
Review,” Current Sustainable/Renewable Energy Reports, vol. 6, pp. 52–60, Jun. 2019, doi: 10.1007/s40518-019-00130-7.
[4] T. R. Ayodele, A. S. O. Ogunjuyigbe, and T. O. Amusan, “Techno-economic analysis of utilizing wind energy for water pumping
in some selected communities of Oyo State, Nigeria,” Renewable and Sustainable Energy Reviews, vol. 91, pp. 335–343, Aug.
2018, doi: 10.1016/j.rser.2018.03.026.
[5] Z. H. Hulio, W. Jiang, and S. Rehman, “Techno - Economic assessment of wind power potential of Hawke’s Bay using Weibull
parameter: A review,” Energy Strategy Reviews, vol. 26, p. 100375, Nov. 2019, doi: 10.1016/j.esr.2019.100375.
[6] Y. Z. Alharthi, M. K. Siddiki, and G. M. Chaudhry, “Resource Assessment and Techno-Economic Analysis of a Grid-Connected
Solar PV-Wind Hybrid System for Different Locations in Saudi Arabia,” Sustainability, vol. 10, no. 10, p. 3690, Oct. 2018, doi:
10.3390/su10103690.
[7] M. Gul, N. Tai, W. Huang, M. H. Nadeem, and M. Yu, “Assessment of Wind Power Potential and Economic Analysis at
Hyderabad in Pakistan: Powering to Local Communities Using Wind Power,” Sustainability, vol. 11, no. 5, p. 1391, Mar. 2019,
doi: 10.3390/su11051391.
[8] W. Sawadogo et al., “Current and future potential of solar and wind energy over Africa using the RegCM4 CORDEX-CORE
ensemble,” Climate Dynamics, vol. 57, pp. 1647–1672, Sep. 2021, doi: 10.1007/s00382-020-05377-1.
[9] A. I. Idriss, R. A. Ahmed, A. I. Omar, R. K. Said, and T. C. Akinci, “Wind energy potential and micro-turbine performance
analysis in Djibouti-city, Djibouti,” Engineering Science and Technology, an International Journal, vol. 23, no. 1, pp. 65–70,
Feb. 2020, doi: 10.1016/j.jestch.2019.06.004.
[10] N. M.-A. Mutombo and B. P. Numbi, “Assessment of renewable energy potential in Kwazulu-Natal province, South Africa,”
Energy Reports, vol. 5, pp. 874–881, Nov. 2019, doi: 10.1016/j.egyr.2019.07.003.
[11] A. Sundaram, A. A. Mas’ud, H. Z. Al Garni, and S. Adewusi, “Assessment of off-shore wind turbines for application in Saudi
Arabia,” International Journal of Electrical and Computer Engineering (IJECE), vol. 10, no. 5, pp. 4507–4513, Oct. 2020, doi:
10.11591/ijece.v10i5.pp4507-4513.
[12] M. K. Abd, “Economic viability and profitability assessments of WECS,” International Journal of Electrical and Computer
Engineering (IJECE), vol. 10, no. 2, pp. 1220–1228, Apr. 2020, doi: 10.11591/ijece.v10i2.pp1220-1228.
[13] S. S. Saswat, S. Patra, D. P. Mishra, S. R. Salkuti, and R. N. Senapati, “Harnessing wind and solar PV system to build hybrid
power system,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 12, no. 4, pp. 2160–2168, Dec.
2021, doi: 10.11591/ijpeds.v12.i4.pp2160-2168.
[14] A. I. Idriss, R. A. Ahmed, A. I. Omar, H. A. Atteyeh, M. H. Ibrahim, and T. C. Akinci, “Wind and solar energy potential in
Herkalou and Lake Assal locations, Djibouti,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 14,
Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss)
100  ISSN: 2252-8792

no. 1, pp. 461–470, Mar. 2023, doi: 10.11591/ijpeds.v14.i1.pp461-470.

[15] H. J. Queen, J. Jayakumar, and T. J. Deepika, “Comparative techno-economic analysis of power system with and without
renewable energy sources,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 24, no. 3, pp. 1260–1268,
Dec. 2021, doi: 10.11591/ijeecs.v24.i3.pp1260-1268.
[16] A. Sakhrieh, J. Al Asfar, and N. A. Shuaib, “An optimized off-grid hybrid system for power generation in rural areas,”
International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 13, no. 2, pp. 865–872, Jun. 2022, doi:
10.11591/ijpeds.v13.i2.pp865-872.
[17] M. A. Abdelrahman, R. H. Abdel-Hamid, M. A. Abo Adma, and M. Daowd, “Techno-economic analysis to develop the first wind
farm in the Egyptian western desert at Elkharga Oasis,” Clean Energy, vol. 6, no. 1, pp. 211–225, Feb. 2022, doi: 10.1093/ce/zkac006.
[18] M. Daoudi et al., “Techno-economic feasibility of two onshore wind farms located in the desert area: a case study,” Energy
Sources, Part A: Recovery, Utilization, and Environmental Effects, vol. 45, no. 3, pp. 7544–7559, Aug. 2023, doi:
10.1080/15567036.2023.2222679.
[19] O. A. Dabar, M. O. Awaleh, D. Kirk-Davidoff, J. Olauson, L. Söder, and S. I. Awaleh, “Wind resource assessment and economic
analysis for electricity generation in three locations of the Republic of Djibouti,” Energy, vol. 185, pp. 884–894, Oct. 2019, doi:
10.1016/j.energy.2019.07.107.
[20] R. Rathi, C. Prakash, S. Singh, G. Krolczyk, and C. I. Pruncu, “Measurement and analysis of wind energy potential using fuzzy
based hybrid MADM approach,” Energy Reports, vol. 6, pp. 228–237, Nov. 2020, doi: 10.1016/j.egyr.2019.12.026.
[21] M. A. Saeed, Z. Ahmed, J. Yang, and W. Zhang, “An optimal approach of wind power assessment using Chebyshev metric for
determining the Weibull distribution parameters,” Sustainable Energy Technologies and Assessments, vol. 37, p. 100612, Feb.
2020, doi: 10.1016/j.seta.2019.100612.
[22] R. Blažević, H. Hadžiahmetović, and I. Ahmović, “Statistical Analysis and Assessment of Wind Energy Potential in Sarajevo,
Bosnia and Herzegovina,” Tehnicki vjesnik - Technical Gazette, vol. 28, no. 5, pp. 1511–1518, Oct. 2021, doi: 10.17559/TV-
20200720163605.
[23] S. A. Akdağ and Ö. Güler, “A novel energy pattern factor method for wind speed distribution parameter estimation,” Energy
Conversion and Management, vol. 106, pp. 1124–1133, Dec. 2015, doi: 10.1016/j.enconman.2015.10.042.
[24] A. H. Shaban, A. K. Resen, and N. Bassil, “Weibull parameters evaluation by different methods for windmills farms,” Energy
Reports, vol. 6, pp. 188–199, Feb. 2020, doi: 10.1016/j.egyr.2019.10.037.
[25] O. M. Salim, H. T. Dorrah, and M. A. Hassan, “Wind speed estimation based on a novel multivariate Weibull distribution,” IET
Renewable Power Generation, vol. 13, no. 15, pp. 2762–2773, Nov. 2019, doi: 10.1049/iet-rpg.2019.0313.
[26] B. A. Çakmakçı and E. Hüner, “Evaluation of wind energy potential: a case study,” Energy Sources, Part A: Recovery,
Utilization, and Environmental Effects, vol. 44, no. 1, pp. 834–852, Mar. 2022, doi: 10.1080/15567036.2020.1811810.
[27] Q. Deltenre, T. De Troyer, and M. C. Runacres, “Techno‐economic comparison of rooftop‐mounted PVs and small wind turbines:
a case study for Brussels,” IET Renewable Power Generation, vol. 13, no. 16, pp. 3142–3150, Dec. 2019, doi: 10.1049/iet-
rpg.2019.0032.
[28] A. Gungor, M. Gokcek, H. Uçar, E. Arabacı, and A. Akyüz, “Analysis of wind energy potential and Weibull parameter estimation
methods: a case study from Turkey,” International Journal of Environmental Science and Technology, vol. 17, pp. 1011–1020,
Feb. 2020, doi: 10.1007/s13762-019-02566-2.
[29] X.-Y. An, Z. Yan, and J.-M. Jia, “A new distribution for modeling wind speed characteristics and evaluating wind power potential
in Xinjiang, China,” Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, May 2020, doi:
10.1080/15567036.2020.1758250.
[30] İ. Mert, F. Üneş, C. Karakuş, and D. Joksimovic, “Estimation of wind energy power using different artificial intelligence
techniques and empirical equations,” Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, vol. 43, no. 7, pp.
815–828, Apr. 2021, doi: 10.1080/15567036.2019.1632981.
[31] C. S. Ambaye, K. T. Beketie, D. Y. Ayal, and Z. T. Dame, “Multiple criteria application in determining wind power potential: A
case study of Adama Zuria woreda, Ethiopia,” Scientific African, vol. 14, p. e01045, Nov. 2021, doi: 10.1016/j.sciaf.2021.e01045.
[32] M. M. Alayat, Y. Kassem, and H. Çamur, “Assessment of Wind Energy Potential as a Power Generation Source: A Case Study of
Eight Selected Locations in Northern Cyprus,” Energies, vol. 11, no. 10, p. 2697, Oct. 2018, doi: 10.3390/en11102697.
[33] J. F. N. Touafio, O. Sanda, S. Malenguinza, J. M’Boliguipa, and R. M. Mouangue, “Analysis of a wind turbine project in the city
of Bouar (Central African Republic),” Scientific African, vol. 8, p. e00354, Jul. 2020, doi: 10.1016/j.sciaf.2020.e00354.
[34] Bangue Centrale De Djibouti, “Bangue Centrale De Djibouti Rapport Annuel 2018,” 2018. [Online]. Available: https://banque-
centrale.dj/wp-content/uploads/2019/10/BCD-Rapport-Annuel-2018_Proof-3.pdf.
[35] M. Irwanto, N. Gomesh, M. R. Mamat, and Y. M. Yusoff, “Assessment of wind power generation potential in Perlis, Malaysia,”
Renewable and Sustainable Energy Reviews, vol. 38, pp. 296–308, Oct. 2014, doi: 10.1016/j.rser.2014.05.075.
[36] A. N. Celik, “A Techno-Economic Analysis of Wind Energy in Southern Turkey,” International Journal of Green Energy, vol. 4,
no. 3, pp. 233–247, May 2007, doi: 10.1080/15435070701338358.

BIOGRAPHIES OF AUTHORS

Abdoulkader Ibrahim Idriss is an assistant professor at Faculty of Electrical and

Energy Engineering, Université de Djibouti, Djibouti. He holds a Ph.D. degree in Photonic
Engineering in France (Université de Franche-Comté, Besançon) with a specialization in
optical nano-antennas for the inspection of photonic structures. His research areas are
materials, photonics, and nanomaterials for renewable energy. He is Dean of the Faculty
of Engineering. He was also the Director of the Logistic and Transport Centre
(Centre of Excellence), financed by the World Bank, from 2019 to December 2021.
He is currently a professor in the department of electrical engineering. He is also Guest
Editor of the Special Issue Big Data in Renewable Energy for Renewable and
Sustainable Energy Reviews published by Elsevier’s Ltd. He can be contacted at email:
[email protected].

Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101
Int J Appl Power Eng ISSN: 2252-8792  101

Ramadan Ali Ahmed received a Ph.D. in nanoscience’s. He completed his Ph.D.

in Laboratory of Molecular Physics at the Faculty of Science and Technic at the Université de
Franche-Comté (France) in 2010. He is assistant-professor at the Faculty of Science since
2002 and also gives courses at the Faculty of Engineering in electricity and magnetism. Dean
of the faculty of science since 2012 at the Université de Djibouti, he is currently Dean of the
Institute of Industrial Technology. His current research is focused on the fields of renewable
energies, photonics, and mechanical engineering. He can be contacted at email:
[email protected].

Hamda Abdi Atteyeh obtained her Bachelor of Engineering (Electrical) and

M.Sc. in Electrical and Energy from the Faculty of Engineering at the Université de Djibouti.
Her research interests include electrical power, electrical distribution systems, system
engineering, energy efficiency, and system reliability. She can be contacted at email:
[email protected].

Abdou Idris Omar is a lecturer in the Electrical Engineering Department,

Université de Djibouti, Djibouti, since 2015, and he has been a senior lecturer since 2015 as a
Fluid Mechanics and Energetics Engineer. He obtained his engineering’s degree from the
Department of Mechanics-Energetics of the ENSIAME, one of the European "Grandes écoles"
of engineering located in the city of Valenciennes in the north of France. He is currently Ph.D.
in fluid mechanics and hydraulics in the Department of Energy System of the Faculty of
Engineers at the Université de Djibouti, Djibouti. His main areas of research interest are
energy efficiency in buildings, energy design, thermal comfort, bioclimatic design,
applications of computational fluid dynamics, and renewable energy. He can be contacted at
email: [email protected].

Tahir Cetin Akinci received B.S. degrees in Electrical Engineering, M.S.

degrees, and Ph.D. degrees from Marmara University, respectively in 2000, 2003, and 2009.
His research interests include electrical power systems, power electronics, computer
programming, artificial neural network, machine learning, and signal processing technique. It
is also actively researching in the Istanbul Technical University (ITU) cognitive system lab.
He has been a visiting professor at the University of California Riverside (UCR), WCGEC
since 2021. He has been working as a professor in the Electrical Engineering Department of
Istanbul Technical University, Turkey. He can be contacted at email: [email protected].

Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss)