Technical Assessment of a Photovoltaic Panel and a Wind Domestic

Turbine Systems in Morocco

A. Serbouti , M. Rattal, A. Boulal, E. Oualim and Az. Mouhsen

Laboratory of Radiation - Matter & Instrumentation, University Hassan First

Faculty of Science and Technology Settat, Morocco

Keywords: Photovoltaics, Wind Power, Optimization, Sensitivity Analysis, TRNSYS

Abstract: This paper presents a general methodology of optimizing the energy performance of a photovoltaic panel in

five different cities in Morocco, by varying its slope and azimuth. A domestic wind turbine is also studied in

Casablanca. For the same capacity, simulations in TRNSYS software prove that photovoltaic panels have a

better yield than the domestic wind turbines. These wind generators can be coupled with the photovoltaic

system in order to lessen the intermittence of the photovoltaic production at night and during short cloudy

days.

1 INTRODUCTION

Morocco benefits from a remarkable potential in

renewable energies thanks to its suitable

geographical position. In fact, our national rate of

sunshine is the ninth best in the world: Morocco’s

710,000-km2 lands profit from a range of 2800 and

3400 hours of sunshine per year. The Moroccan

Agency For Solar Energy (MASEN) evaluates the

Moroccan technical potential of solar energy to

20,000 MW. However, the actual installed capacity

is only estimated to 180 MW in 2016 (MASEN,

2018).

Among the various available technical

solutions to exploit the energy of the sun,

photovoltaic technologies make it possible to

convert the sunlight into electricity, and global

efficiencies are typically around 14-16% for

polycrystalline modules.

Many Moroccan authors studied the

performance of PV installations; K. Attari and al.

presented an evaluation of a grid-connected

photovoltaic (PV) system installed on the roof of a

government building located in Tangier,

Morocco(Attari, 2016). D. Lahjouji and al.

optimized the tilt angle for maximum solar energy

collection in Ifrane, Morocco (Lahjouji, 2013).

In the framework of the study, a PV system is

modelled in TRNSYS transient simulation program

using a PV panel (Type 94) and a typical

meteorological year (TMY2) conditions (with

Meteonorm software data). The panel performance

is studied and optimized in five different cities in

Morocco: Casablanca, Fez, Tangier, Ouarzazate and

Marrakech, using genetic algorithms. These cities

are located in five different climatic zones according

to the Moroccan thermal regulation for buildings

(ADEREE, 2015).

The global installed capacity of wind turbines

was about 318,1 GW by the end of 2013. Morocco

has launched, on June 28, 2010 an ambitious wind

energy program, aiming to grow the wind plants to

2000 MW by 2020. The commissioning of the first

wind farm in Morocco took place in 2000

(Abdelkhalek Torres Farm in Tetouan, 50.4 MW)

(ATLAS ADEREE, 2018). Many other wind energy

generation projects were completed since then:

Amougdoul in Essaouira (60 MW), Tangier wind

farm (140 MW), Houma (50 MW)…

Mohamed Oukili and al. performed a

comparative Study of the Moroccan Power Grid

Reliability in Presence of Photovoltaic and Wind

Generation. They conclude that wind and solar

power sources to be used in order to save fossil fuel

and increase the total energy generation in Morocco

(Oukili, 2013).

Serbouti, A., Rattal, M., Boulal, A., Oualim, E. and Mouhsen, A.

Technical Assessment of a Photovoltaic Panel and a Wind Domestic Turbine Systems in Morocco.

DOI: 10.5220/0009774902110217

In Proceedings of the 1st International Conference of Computer Science and Renewable Energies (ICCSRE 2018), pages 211-217

ISBN: 978-989-758-431-2

211

Consequently, the second chapter of this article

will shed the light on the study of a domestic wind

turbine, with a capacity of 1Kw in Casablanca, using

the type 90 of TRNSYS and the weather data (wind

and direction speed) collected by Meteonorm in a

weather station located in Casablanca.

2 STUDY AND OPTIMIZATION

OF THE ENERGY

PERFORMANCE OF A PV

PANEL

2.1 Weather Data

Meteonorm7 software generates the weather data

used in this paper.

Meteonorm is a complete, worldwide

climatological database. The software enables data

generation of hourly values for any place in the

world (Meteonorm, 2018). The user can synthesize

these data in an output file compatible with

TRNSYS software.

For Casablanca, Tangier, Fez, Ouarzazate and

Marrakech, radiation data cover respectively the

period 1991-2010. Figures 1 to 5 show the global

and diffuse radiations evolution in the five cities.

Figure 1: Global (yellow) & Diffuse (orange)

radiations in Casablanca;

Figure 2: Global (yellow) & Diffuse (orange)

radiations in Fez;

Figure 3: Global (yellow) & Diffuse (orange)

radiations in Marrakech;

Figure 4: Global (yellow) & Diffuse (orange)

radiations in Tangier;

Figure 5: Global (yellow) & Diffuse (orange)

radiations in Ouarzazate;

These cities and Morocco in general, benefit

from a very important global radiation (255 kWh/m²

per month at its highest in Ouarzazate). Morocco is

then an attractive country for investing in solar

technologies.

2.2 TRNSYS Model and Initial

Results

In our case study, we chose a photovoltaic panel

produced locally in Skhirat, Morocco by PV

Industry, subsidiary company of Jet Energy and

specialized in Photovoltaics.

ICCSRE 2018 - International Conference of Computer Science and Renewable Energies

212

These panels have a surface of 1,94m² (1,952 x

0,992 m²) and a 300Wc per-unit power. The

TRNSYS model is showed in the figure 7.

Figure 6: PV Industry 300Wc polycrystalline module

parameters in TRNSYS

Figure 7: TRNSYS model

We will study the performance of a single

photovoltaic panel in the five cities, with an initial

slope and azimuth of 0° (horizontal panel, south-

faced). The results are presented in the figure 8.

Figure 8: Energy output in kWh in the five cities

In the following paragraph, we will use a

sensitivity analysis approach to determine which of

the two studied parameters (slope and azimuth) has

the most influence on the energy performance. Then,

we will optimize it by studying them in their

variation intervals.

2.3 Sensitivity Analysis

Systems simulated on dynamic thermal simulation

(DTS) tools present many input parameters. In order

to optimize a chosen output of these tools according

to the combinations made up of input parameters,

sensitivity analysis allows the identification of the

parameter or set of parameters that have the greatest

influence on the model output, and thus not to study

the parameters which have a low influence on the

model (École Chercheur Mexico, 2010). Sensitivity

analysis helps determining how a digital model

answers variations intervening on its inputs (Looss,

2011).

Sobol sensitivity analysis determines the

contribution of each input parameter and their

interactions to the overall model output variance.

The sensitivity of the output compared to the

parameters is given by various orders indices of

sensitivity.

Among these indices, the total order index allows

to study both the effect of the parameter alone and

the effects of its interaction with all the other

parameters on the variation of the output.

We developed algorithms on the OpenSource

programming language Python, in order to adapt our

case study to the algorithms of the SALIB library

(Sensitivity Analysis Library) available on GITHUB

(Herman, 2017).

The studied parameters are:

• Slope (from 0 horizontal to 90° vertical);

• Azimuth (from 0 south to 360° south);

Figure 9: Sensitivity analysis first and total indexes

We notice that the two parameters have almost

the same first order index (same influence when

varying one parameter and fixing the other).

The total index of the azimuth is although higher

than the total index of the collector slope. The panel

azimuth has then a slightly higher influence on the

energy yield than the panel slope.

The second order index shows also that there is a

high interaction between these two parameters.

Technical Assessment of a Photovoltaic Panel and a Wind Domestic Turbine Systems in Morocco

213

2.4 Optimization using MOBO

Software

In this paragraph, we carry out a coupling of the

TRNSYS software with MOBO (Multi-Objective

Building Optimization tool) optimization tool,

developed by the Technical Research Centre of

Finland (Palonen, 2013).

To perform the optimizations, we will use the

genetic algorithms GA (in specific the non-

dominated sorting genetic algorithm NSGA-II),

coupled to the Hooke-Jeeves algorithm (Deb, 2001).

GA reflect the process of natural selection where the

fittest individuals are selected for reproduction in

order to produce offspring of the next generation

(Dubrow, 2010). The Hooke-Jeeves algorithms

perform modified iterations of Hooke and Jeeves

until no further progress is forthcoming.

Figure 10: Coupling TRNSYS / MOBO GA

The table below shows the optimization results

in the five cities.

Table 1 - Optimization results

City

Azimuth

(°)

Slope

(°)

Optimized

Yield

(kWh)

Initial

Yield

(kWh)

Casablanca 0 29,52 629 564

Fez 0 31,21 663 587

Tangier 0 31,97 690 604

Marrakech 358,4 30,32 704 627

Ouarzazate 0,05 30,81 765 678

We notice that in the five cities, the optimal

orientation of the panel is south (0°), and the optimal

slope is around 30°.

By optimizing these two parameters, we gain

around 12% of the annual electricity production.

Fig. 11: Optimization process (510 simulations) of the

energy yield inCasablanca

2.5 Comparison with Random

Search Algorithm

The random search algorithm randomly varies the

values of the two parameters within their intervals of

variation, until reaching the number of simulations

specified by the user.

With 500 simulations, the optimal combination

is:

- Yield : 627 kWh;

- Azimuth : 355°;

- Slope : 29.55°;

Figure 12: Random search algorithm

We notice that the random search algorithm

permits having a good approximation of the optimal

value (627 kWh instead of 629 kWh), although the

NSGA-II algorithm is more efficient.

2.6 Effect of the Temperature on the

Yield

We studied below the effect of variation of the

temperature (from -5°C to 45°C) on the electrical

output of the most efficient hour of the year. The

figure 13 shows the results obtained with the random

search algorithm.

We notice that the yield is decreasing while

increasing the ambient temperature. The optimal

operating temperature is 5°C.

1000

103

137

171

205

239

273

307

341

375

409

443

477

ICCSRE 2018 - International Conference of Computer Science and Renewable Energies

214

The PV panel is losing around 8% of its

efficiency when passing from 5 to 45°C.

Figure 13: Effect of the temperature on the electrical

yield of 1 hour (in Watt)

3 DOMESTIC WIND TURBINE

STUDY

Solar energy is generally not enough alone to supply

the needed energy, especially during the night and in

the short gray days of winter. Moreover, according

to local weather statistics, wind speed increases in

winter, so that solar energy and wind energy can

complement each other. In addition, small wind

turbines have a rather simple manufacturing

technology and require less land area (Elnaggar,

2017).

For these considerations, we will study in this

section a domestic wind turbine installation in

Casablanca.

3.1 Weather Station

The weather station used in this study is located in

Casablanca, its coordinates are 33,6°N / -7,7°E, 55m.

Figure 14 below shows the wind rose and the wind

speed evolution in Casablanca.

Figure 14: Wind rose in Casablanca

Then, the preponderant wind directions in

Casablanca are NNE, NE, SW, SSW and South.

In the figure 15, Y-axis shows the sum of hours per

year (for a total of 8760 hours per year) and X-axis

refers to the wind speed in m/s.

Figure 15: Wind speed distribution in Casablanca

3.2 Domestic Wind Device and

TRNSYS Model Weather Station

In this paragraph, we will study the

performance of a 1kW wind Turbine (HUMMER

1kW wind generator). The TRNSYS model and

thecharacteristics of the generator are detailed below

(HUMMER, 2018).

Figure 16: TRNSYS model.

Figure 17: HUMMER 1kW Power / wind speed curve.

3.3. Results

The figure 18 shows the hourly wind power

generation in Casablanca (in Watts).

Technical Assessment of a Photovoltaic Panel and a Wind Domestic Turbine Systems in Morocco

215

Figure 18: Hourly cumulative power generation (8760

hours per year)

Figure 19: Hourly power generation (8760 hours per year)

The annual electrical production is then

around 650 kWh, for the 1kW domestic wind

turbine, while a single photovoltaic panel may

produce the same yield as showed above.

We deduce then that for the same capacity,

photovoltaic panels have a better annual electric

yield than the domestic wind turbines in Casablanca.

However, these domestic wind turbines can be

coupled with a photovoltaic installation, in order to

lessen the intermittence of the photovoltaic

production at night and during short cloudy days

(Figure 20).

Figure 20: PV and wind yield in the 31th of December

4 CONCLUSIONS

The sunlight harvesting and the photovoltaic

applications are interesting in Morocco, due to the

important global radiations received by its surface.

In the scope of this work, we studied the energy

performance of a PV panel manufactured locally in

Skhirat by PV Industry. The electric output was

optimized following the intervals of variation of the

panel slope and azimuth. It was proved that a south-

facing panel, with a slope of around 30° gives the

best annual energy yield. A sensitivity analysis

performed in the city of Casablanca, shows that the

azimuth influences more the annual yield than the

slope of the panel.

Furthermore, Meteonorm collected weather

data in the Meteonorm station of Casablanca (wind

speed and direction) were used to study a domestic

wind farm of 1kW. The simulation gives an annual

production of around 650 kWh, which is almost the

production of a single-unit photovoltaic panel. For

the same capacity, photovoltaic panels have than a

better yield then the domestic wind turbines in

Casablanca. These two technologies can be coupled

in order to lessen the intermittence of the

photovoltaic production at night and during cloudy

days.

REFERENCES

Kamal Attari, Ali El Yaakoubi, Adel Asselman,

Performance analysis and investigation of a grid-

connected photovoltaic installation in

Morocco,Energy Reports Volume 2, November 2016,

Pages 261-266

Driss Lahjouji, Hassane Darhmaoui, Tilt angle

optimization for maximum solar energy collection -

Case study for Ifrane, Morocco, 2013 International

Renewable and Sustainable Energy Conference

(IRSEC)

Mohamed Oukili, Smail Zouggar, Mohamed Seddik,

Taoufik Ouchbel,François Vallée, Mohamed El

Hafiani, Comparative Study of the Moroccan Power

Grid Reliability in Presence of Photovoltaic and Wind

Generation, Smart Grid and Renewable Energy, 2013,

4, 366-377 http://dx.doi.org/10.4236/sgre.2013.44043

MASEN OU LE DÉVELOPPEMENT

RENOUVELABLE, Moroccan Agency For Solar

Energy

Agence Nationale pour le développement des Energies

Renouvelables et de l’efficacité énergétique

(ADEREE), Règlement thermique de construction au

Maroc (RTCM), 2015

ICCSRE 2018 - International Conference of Computer Science and Renewable Energies

216

Agence Nationale pour le Développement des Energies

Renouvelables et de l’Efficacité Energétique - ATLAS

ÉOLIEN GLOBAL NUMÉRIQUE DU MAROC

Meteonorm Software

Looss, Revue sur l’analyse de sensibilité globale de

modèles numériques, 2011

École Chercheur Mexico – Analyse de sensibilité: mesure

de l’importance des facteurs par décomposition de la

variance 9 Juin 2010

J. Herman, W. Usher - SALib Documentation – October

2017

MattiPalonen, Mohamed Hamdy, AlaHasan - MOBO A

NEW SOFTWARE FOR MULTI-OBJECTIVE

BUILDING PERFORMANCE OPTIMIZATION -

Technical Research Centre of Finland, Espoo, Finland

– 2013

Deb, K. Multi-Objective Optimization using evolutionary

algorithms; John Wiley & Sons: Chichester,UK, 2001

Tuhus-Dubrow D, Krarti M. Genetic-algorithm based

approach to optimize building envelope design for

residential buildings. Build Environ 2010; 45:1574–81

Mohamed Elnaggar, EzzaldeenEdwan and Matthias Ritter

2 Wind Energy Potential of Gaza Using Small Wind

Turbines: A Feasibility Study, Department of

Engineering, Palestine Technical College, 18 August

2017

HUMMER 1kW wind generator technical datasheet

Technical Assessment of a Photovoltaic Panel and a Wind Domestic Turbine Systems in Morocco

217