Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 93 (2016) 74 – 81

Africa-EU Renewable Energy Research and Innovation Symposium, RERIS 2016, 8-10 March
2016, Tlemcen, Algeria

Wind energy resource assessment in Ngaoundere locality

Myrin Y. Kazeta,b*, Ruben Mouangueb, Alexis Kuitchea, J.M. Ndjakac
a
Department of GEEA, PAI, ENSAI, University of Ngaoundere, PO Box 455, Ngaoundere, Cameroon
b
Department of Energetic Engineering, UIT, UN, PO Box 455, Ngaoundere, Cameroon
c
Department of Physics, Faculty of Sciences, University of Yaounde I, PO Box 812, Yaounde, Cameroon

Abstract

In order to secure future energy and protect the environment, it is important to consider the possibilities of wind as a resource for
electrical energy supply. To carry out this study in Cameroon, we chose the locality of Ngaoundere, in which an assessment of
the wind energy resource was made. Different kinds of data have been collected about climate, topography, and roughness. The
Observed Wind Climate of the meteorological station has been made. The Wind Atlas and the resource grid have been
calculated, especially in the high wind resource areas. Annual Energy Production of one hypothetical wind farm consisting of
four 1.65 MW turbines was estimated using the Weibull-representative wind data for a total of 12 months. The computed Annual
Energy Production is 5,985 MWh and according to the International Energy Agency statistics, this production could enable the
reduction of CO2 emission by 1200 tons per year. It was found that there is a good correlation between our calculation results and
those of the Wind Atlas Analysis and Application Program (WAsP).

©
© 2016
2016TheTheAuthors. Published
Authors. by by
Published Elsevier Ltd.Ltd.
Elsevier This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of RERIS 2016.
Peer-review under responsibility of the organizing committee of RERIS 2016
Keywords: Weibull distribution; resource assessment; wind energy; wind potential; Ngaoundere.

1.
Introduction

Recent years have witnessed a fundamental change in the way governments approach energy-related
environmental issues. Promoting sustainable development and combating climate change have become integral

* Corresponding author. Tel.: +237 699 97 03 39 / +237 677 46 10 06.

E-mail address: myrin_kz@yahoo.fr

1876-6102 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of RERIS 2016
doi:10.1016/j.egypro.2016.07.152
 Myrin Y. Kazet et al. / Energy Procedia 93 (2016) 74 – 81 75

Fig. 1. (a) Topographic map of the Ngaoundere locality; (b) an aerial view of the site; (c) the map of modelled obstacles.

aspects of energy planning, analysis and policy making in many countries. In Cameroon, the question of
development and environment are at the heart of the energy transition. The wind energy, which confirmed its status
as the number two source of renewable electricity production in 2012, is now the most likely renewable energy
source to back up the hydropower supply in halting the relentless increase in fossil fuel used to generate power [1].
However, this energy remains unexploited in Cameroon [2] in spite of its theoretical potential in the North of the
country. The aim of this paper is to carry out the wind characteristic of the Ngaoundere locality and furthermore
provide a wind resource map useful for the selection of suitable areas for wind park installation. It is a technical
study that aims at bringing an efficient help to all decision makers, with respect to the planning and realization of
wind energy projects. For this, we have collected and processed the data of our site. Then, using the models we have
validated, we performed an assessment of the energy potential of this site.
76 Myrin Y. Kazet et al. / Energy Procedia 93 (2016) 74 – 81

2.
Methods

2.1.
Collection and processing of data

2.1.1.
Wind
Data about wind used in this work were obtained at the meteorological airport station of Ngaoundere. These wind
data were recorded every day at 30 minute intervals during a 12-month period at the standard height of 10 m above
the ground level.

2.1.2.
Topography
The topography data of our study area were collected using Google Earth in order to create a xyz coordinate file.
Then, this file was interpreted using Golden Surfer and our topographic map was generated (Fig. 1 (a)).

2.1.3.
Obstacle and roughness

From several meticulous visits and aerial view of our site, we have identified and taken into account the main
terrain roughness and obstacles around the site of data collection (Fig. 1 (b)-(c)).

2.2.
Modelling of the observed wind speed distribution

The probability density function f(v) indicates the fraction of time for which the wind is at a given speed v
[3,4,5].

: = :&! = :
  P
 O (1)
- - -

The cumulative distribution function F(v) gives the fraction of time that the wind speed is equal or lower than v
[6].

= :
  P
 O  O
(2)
-

According to Seguro and Lambert (2000) [7], when the wind speed data are available in the frequency
distribution format, Weibull parameters are estimated using Eqs. (3) - (4).

L K L &!
JEA =J 2; =J 8Q=J R JEA 2; =J 8Q=J R
 P
L = K 8Q= R O

(3)
JEA J J 0Q=, R

A
! ; : K
 P
9'! 9  9


(4)
0 =,

In this work, Eq. (3) is solved using an iterative procedure with a Fortran 90 code. The initial guess is k = 2 [7].
After the convergence of the k value, Eq. (4) is solved explicitly.
The mean wind speed and the power density, following the Weibull distribution, is given by Eq. (5)-(6) [4, 6].

!

P


  N (5)
:

! #
. P
 #   N (6)
" :

The energy pattern factor EPF is given by Eq. (7) below [4]:
 Myrin Y. Kazet et al. / Energy Procedia 93 (2016) 74 – 81 77

C
6C > !%

K
30 P

P
A (7)
6C > C !%

K

2.3.
Modelling of the vertical extrapolation of the wind speed

The Hellmann exponential law was used since it was found to give a reasonably accurate representation of wind
speed profiles and can be used to adjust the data reasonably well from 10 m up to 150 m above ground level [8, 9,
10].

1 ?
P
4 (8)
1G

with

+ ) (

P
(* ) )F & ) (
(9)

where Z̅ is the geometric mean height of observations at HR and H:

!
 P Q  4 R " (10)

Our site has a moderate relief and is tree covered. By using the table of roughness length of different types of
surface given by McRae et al. (1982) [11], we estimated the roughness length to be ZO = 1.0 m. Also, by using the
Pasquill dispersion classes [12, 13] we determined that the class stability of our site is B (unstable). Hence, ϕ and ψ
stability functions were selected [9].

&! $
 
P
 O  
(11)

@MB %! @M %! B
 
P
O


O    O  < (12)
@ B %! @%! B

with:

! $
 P
 O  
(13)

! $
< P
 O  <
(14)

The inverse of the Monin – Obukhov length 1/L can be calculated as a function of Z0 [9, 12].

7

P  < (15)

where coefficients a and b are tabulated in [12] as a function of the stability class.

2.4.
Modelling of the power curve

Actual power curves are rather smooth and can be well approximated by a piece-wise linear function with a few
nodes (Pi, vi). The mean power production PEL based on a Weibull probability density function can be expressed by
Eq. (16) [14].
78 Myrin Y. Kazet et al. / Energy Procedia 93 (2016) 74 – 81

  













    
  

Fig. 2. Weibull probability density function compared with the observed wind speed frequency histogram.

6L 6H
/2 P

6I 5

   N
; 6L
   (16)

3JDA &3J
5  P  O 9 N 9 (17)
=JDA &=J

Hence, the annual energy production is

/2 P
/2  1 (18)

NH is the time period of data collection; its value is 8784 hours (leap year).

3.
Results and discussion

Weibull parameters, the mean speed at 10 m, the energy density, the mean wind speed at 100 m and the Annual
Energy Production (AEP) have been calculated by our Fortran code and the results presented in Table 1 are
compared to those obtained by WAsP.


The values of Weibull parameters calculated by the modified maximum likelihood method are very close to those
obtained by WAsP [15]. An agreement is also observed on the mean wind speed and the energy density calculated at
10 m.
After values of k and C had been determined, it was now possible that the Weibull distribution be fitted to the
observed frequency of the wind speed obtained from data. Fig. 2 shows the wind speed Probability Density Function
(PDF) of Weibull with parameters determined by the modified maximum likelihood method. The solid line in blue
represents the Weibull PDF while the black line represents the observed wind speed frequency histogram. It can be
seen in this figure that the PDF curve seems to match well the histogram.

In order to select suitable areas for a wind farm, we have performed a wind resource extrapolation of our site
using for this task, the topographic map that we have generated and the WAsP software.

Table 1. Weibull parameters, mean wind speed, mean energy density and AEP at the meteorological station of Ngaoundere.

k C (m/s) V̅ 10 m (m/s) ED10m (W/m2) EPF V̅ 100 m (m/s) AEP (MWh)

Our results 1.236274 1.690723 1.578156 7.593257 3.767907 3.618266 1226.3
From WAsP 1.230 1.700 1.570 8.00 - 3.580 1257.0
 Myrin Y. Kazet et al. / Energy Procedia 93 (2016) 74 – 81 79

Fig. 3. Maps of the mean wind speed over our site in Ngaoundere at 100 m height. On (a), the wind rose indicates the meteorological station
where data were collected and the zones in red are expected to have a high wind resource. On (b) the high wind resource zones are coloured in
blue.

One of the outputs consists of predictions of the wind speed distribution (Figs. 3(a)-(b)) over the whole area. For
each node of the map, the wind speed was computed knowing the height Z. The results of the computation are
showed in Fig. 3. The map (b) is a 3D plot of the wind speed distribution and was carried out by using the Golden
software.
Zones with high wind resource are coloured in red. As expected, the resource is relatively high along the crest of
the hills and in particular over the big hill to the East of the data collection site. However, the resource is low around
the data collection site and lower over zones with small elevations in the northern and north-eastern parts of the
resource map area. In the next step, we decided to explore a particular zone of higher wind resource: the hill at the
East of data collection site. Its localization coordinates are UTM 33 387,000 m – 391,680 m E and 810,500 m –
815,180 m N.

One hypothetical wind farm (Fig. 5) consisting of four 1.65 MW turbines (Vestas V82) was simulated on this
particular zone. After computations, the output consists of predictions of the mean wind speed, the total energy
produced (Gross), the AEP (Net) and the wake effects (Loss).

Fig. 4. The Projection of the selected area on the Earth.

80 Myrin Y. Kazet et al. / Energy Procedia 93 (2016) 74 – 81

Fig. 5. The selected area and the hypothetical wind farm at the East of Ngaoundere town.

The turbine height is 100 m. Table 2 shows the statistics for each wind turbine of the wind farm.
For the particular zone with higher wind resource that we choose, the AEP of a hypothetical wind farm consisting
of four turbines of 1.65 MW was 5,985 MWh.
According to the International Energy Agency statistics, CO2 emissions per kWh from electricity generation in
Cameroon are 200 grams [16]. The production of such wind farms could thus enable to reduce the CO2 emissions by
1200 tons per year.

Table 2. The wind turbine siting and their annual energy production.

Site description X-location [m] Y-location [m] Elev. [m] U [m/s] Gross[GWh] Net [GWh] Loss [%]
Turbine site 001 388920.6 813708.4 1500 3.87 1.485 1.469 1.08
Turbine site 002 388629.5 814322.1 1500 3.93 1.548 1.532 1.04
Turbine site 003 387857.5 814366.2 1500 3.97 1.576 1.549 1.72
Turbine site 004 388134.0 813801.4 1500 3.89 1.501 1.435 4.40

4.
Conclusions

In this study, we aimed to provide a wind resource map and a technical study concerning the Ngaoundere
locality. Also, in order to model useful physical parameters for the wind potential assessment, we used some models
and our computing results were compared with those of WAsP.
From the wind atlas generated, we elaborated the wind resource card of the locality of Ngaoundere. On this card,
we selected one zone of higher wind resource in which it is assumed that there is one hypothetical wind farm,
consisting of four 1.65 MW turbines in appropriate configurations. AEP of that wind farm is computed to be 5,985
MWh and it is expected that this production would enable us to avoid the rejection of 1200 tons of CO2 per year in
the environment.
It was also found that the predictions of models that we used compared to WAsP predictions agreed as well.
Furthermore, the modified maximum likelihood method proposed by Seguro and Lambert in 2000, which is
recommended for use with wind data in frequency distribution format, is one of the best methods which can be used
 Myrin Y. Kazet et al. / Energy Procedia 93 (2016) 74 – 81 81

to determinate Weibull parameters. Weibull distribution appears to be fine for the assessment of the wind potential
and the energy output.
On a global level, CO2 emissions are growing with the energy demand. Because it is now an urgent need to
reconcile our development model with the increasing environmental and social stakes, we expect that this work will
contribute to making wind power one of the future pillars for electricity supply in the rural areas in the north of
Cameroon and thus consolidate its implication in the sustainable development.
Research in this area is still in its starting phase in Cameroon. One of the main obstacles is the lack of
infrastructure. A partnership between European and African universities would be fruitful for both parts. Africa
might benefit from the EU expertise and infrastructures and EU in return might receive from Africa his own
experience and availability of researchers.

Acknowledgements

The authors would like to thank the Risø National Laboratory, Technical University of Denmark for the
academic license of WAsP 11 provided. Also, the ASECNA weather service of Ngaoundere for the meteorological
data used in this work.

References

[1] Furkan D. The analysis on wind energy electricity generation status, potential and policies in the world. Ren & Sust Energy Reviews: 15,
5135– 5142, 2011.
[2] Observ’ER, EDF. Electricity production in the world: Fifteenth inventory edition, 2013.
[3] Bataineh KM, Dalalah D. Assessment of wind energy potential for selected areas in Jordan. Ren Energy: 59, 75–81, 2013.
[4] Akdağ SA, Dinler A. A new method to estimate Weibull parameters for wind energy applications, Energy Convers and Management, 50
(2009), 1761–1766.
[5] Safari B, Gasore J. A statistical investigation of wind characteristics and wind energy potential based on the Weibull and Rayleigh models in
Rwanda, Ren Energy, 35(2010), 2874–2880.
[6] Ohunakin OS, Akinnawonu OO. Assessment of wind energy potential and the economics of wind power generation in Jos, Plateau State,
Nigeria, Energy for Sust Develop, 16(2012), 78–83.
[7] Seguro JV, Lambert TW. Modern estimation of the parameters of the Weibull wind speed distribution for wind energy analysis. J Wind Eng
and Indus Aerodynamics: 85, 75–84, 2000.
[8] Bañuelos-Ruedas F, Angeles-Camacho C and Rios-Marcuello S. Analysis and validation of the methodology used in the extrapolation of
wind speed data at different heights, Ren & Sust Energy Rev, 14 (2010), 2383–2391.
[9] Gualtieri G, Secci S. Methods to extrapolate wind resource to the turbine hub height based on power law: A 1-h wind speed vs. Weibull
distribution extrapolation comparison. Ren Energy: 43, 183-200, 2012.
[10] Sedefian L. On the vertical extrapolation of mean wind power density, J Applied Meteorology, 19 (1980), 488 – 493.
[11] Mcrae GJ, Goodin WR, Seinfeld JH. Development of a second generation mathematical model for urban air pollution, I. Model formulation.
Atmospheric Environment, 1982, 16(4): 679-696.
[12] Zannetti P. Air Pollution Modeling - Theories, Computational Methods and Available Software. Springer Science + Business Media, LLC,
p. 42, 49, 60, New York, 1990.
[13] Seinfeld JH, Pandis SN. Atmospheric Chemistry and Physics – From Air Pollution to climate change, 2nd edition, John Wiley & Sons, Inc.,
p. 750, New Jersey, 2006.
[14] Carta JA, Ramírez P and Velázquez S. Influence of the level of fit of a density probability function to wind-speed data on the WECS mean
power output estimation, Energy Conv & Management, 49 (2008), 2647–2655.
[15] Mouangue RM, Kazet MY, Kuitche A, & Ndjaka JM. Influence of the Determination Methods of K and C Parameters on the Ability of
Weibull Distribution to Suitably Estimate Wind Potential and Electric Energy. Int. Journal of Renew Energy Development, 3(2), 145-154.
http://dx.doi.org/10.14710/ijred.3.2.145-154.
[16] IEA Statistics. CO2 emissions from fuel combustion, highlights”, International Energy Agency, [Accessed 17 April 2015].
http://www.iea.org/publications/freepublications/publication/CO2EmissionsFromFuelCombustionHighlights2013.pdf