Tunisian Republic

Ministry of Higher Education and Scientific Research

University of Tunis El Manar
National Engineering School of Tunis
Civil Engineering Department
Laboratory of Modelisation in Hydraulics and Environment

End of 2nd year project

CFD applications in renewable energy

Wind Power case

Prepared by :

Nour El Houda Nafkha

Supervised by :

M. Hatem Kanfoudi

2nd year Hydraulic and Enviromental Engineering

University Year : 2024/2025

Abstract
This study investigates the performance of a helical wind turbine through a
combined experimental and numerical approach. Wind tunnel tests are conducted
to analyze aerodynamic behavior, power output, and wake dynamics, while high-
fidelity CFD simulations provide detailed insights into airflow characteristics and
turbulence effects. By integrating experimental data with numerical predictions,
this research aims to optimize the design of helical wind turbines, improving their
efficiency and enhancing their applicability in sustainable energy systems.

Keywords :
Wind energy, VAWT (Vertical Axis Wind Turbine), Numerical simulation,
CFD (Computational Fluid Dynamics), Scale model, Similarity, Power coefficient,
Experiment, Helical, Savonius.

Résumé
Cette étude examine les performances d’une éolienne hélicoïdale en combinant
une approche expérimentale et numérique. Des essais en soufflerie sont effectués pour
analyser le comportement aérodynamique, la puissance de sortie et la dynamique du
sillage, tandis que des simulations CFD haute-fidélité fournissent des informations
détaillées sur les caractéristiques de l’écoulement de l’air et les effets de la turbulence.
En intégrant les données expérimentales aux prévisions numériques, cette recherche
vise à optimiser la conception des éoliennes hélicoïdales, en améliorant leur efficacité
et en renforçant leur applicabilité dans les systèmes d’énergie durable.

Mots clés:
Energie éolienne, VAWT, Simulation numérique, CFD, modèle réduit, similitude,
coefficient de puissance, expérience, hélicoïdale, Savonius.

1
Appreciation
I would like to express my sincere gratitude to everyone who contributed to the
completion of this final-year project.
First and foremost, I extend my heartfelt thanks to my supervisor Professor
KANFOUDI Hatem for his availability, insightful guidance, and constant support
throughout this work. His assistance has been crucial in ensuring the success of
this project.
My appreciation goes also to Mrs.ENNOURI Marwa for her help and support
all along this project , her advices and resources have been a great guide during
this research .As well as M.BELLAKHEL Ghazi for his precious guidance and
time devoted to the simulation and understanding the background of the software .
I would also like to express my appreciation to ENIT for the academic environ-
ment and resources provided, as well as to the LMHE (Laboratory of Modelling in
Hydraulics and Environment) for its expertise, research, and valuable support.
A special thanks goes to my family, parents and siblings, for their unwavering
support, patience, and encouragement. Their presence and understanding have
been a true driving force throughout this journey.
Lastly, I would like to thank everyone who, directly or indirectly, contributed
to the progress of this project and helped me reach this milestone.

2
Contents

List of Figures 6

Introduction 1

1 Overview Fundamental Concepts 2

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Energies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 Energetic situation in the world . . . . . . . . . . . . . . . . 2
1.2.2 Fossil energies . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Renewables energies . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.1 Green Power energy . . . . . . . . . . . . . . . . . . . . . . 6
1.3.2 Advantages of renewables energies . . . . . . . . . . . . . . 6
1.3.3 Projects to enhance the use of renewable energies . . . . . . 8
1.3.4 Types of renewables energies . . . . . . . . . . . . . . . . . . 9
1.4 Wind Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.4.1 Main components of a wind turbine: . . . . . . . . . . . . . 12
1.4.2 Types of Wind Turbines: . . . . . . . . . . . . . . . . . . . . 12
1.4.3 HAWT and VAWT . . . . . . . . . . . . . . . . . . . . . . . 13
1.4.4 Total efficiency of a wind turbine : . . . . . . . . . . . . . . 14
1.4.5 Choosing a wind turbine for a site in Tunisia . . . . . . . . . 14
1.5 Aerodynamic Concepts . . . . . . . . . . . . . . . . . . . . . . . . 15
1.5.1 BETZ Theory . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.5.2 Aerodynamic variables: . . . . . . . . . . . . . . . . . . . . . 17
1.5.3 Aerodynamic forces . . . . . . . . . . . . . . . . . . . . . . . 19
1.5.4 Aerodynamic coefficients . . . . . . . . . . . . . . . . . . . . 20
1.6 Dimensional Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.7 Key Parameters in the Savonius Wind Turbine . . . . . . . . . . . . 21
1.8 Buckingham π Theorem . . . . . . . . . . . . . . . . . . . . . . . . 24
1.8.1 Key Dimensionless Groups . . . . . . . . . . . . . . . . . . . 24
1.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

4
2 Software 26
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.2 CFD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.2.1 CFD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.2.2 General equations of incompressible flows . . . . . . . . . . . 27
2.2.3 Turbulance Models . . . . . . . . . . . . . . . . . . . . . . . 32
2.2.4 Numerical Simulation . . . . . . . . . . . . . . . . . . . . . . 35
2.2.5 Modeling Turbulent Flow with ANSYS workbench in 5 Steps
including CFX Project . . . . . . . . . . . . . . . . . . . . . 36
2.2.6 Finite Elements Method . . . . . . . . . . . . . . . . . . . . 37
2.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3 Experimental Study of a Vertical-Axis Wind Turbine (Savonius

Wind Turbine) 38
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.2 Geometric Features of the Helical Savonius . . . . . . . . . . . . . . 38
3.3 Wind Turbines Available in our Laboratory . . . . . . . . . . . . . . 39
3.4 Experimental Study of a Helical Savonius . . . . . . . . . . . . . . . 42
3.4.1 Purpose of the Manipulation . . . . . . . . . . . . . . . . . . 42
3.4.2 Experimental Procedure and Equipment . . . . . . . . . . . 42
3.4.3 Experimental Results . . . . . . . . . . . . . . . . . . . . . . 44
3.4.4 Used Formulas (Mariem Lajnef, 2020) . . . . . . . . . . . . . 44
3.4.5 Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.4.6 Interpretation of the Results . . . . . . . . . . . . . . . . . . 47
3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

4 Numerical Simulation 48
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.1.1 Numerical Study of the Helical Vertical-Axis Wind Turbine . 48
4.1.2 Geometry of the Savonius . . . . . . . . . . . . . . . . . . . 48
4.1.3 The Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.1.4 General Flow Conditions (Set-Up) . . . . . . . . . . . . . . . 51
4.1.5 Calculation Convergence . . . . . . . . . . . . . . . . . . . . 51
4.1.6 Results and Discussion . . . . . . . . . . . . . . . . . . . 53
4.2 Comparison Between Numerical and Experimental Results . . . . . 56
4.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Bibliographic References 59

5
List of Figures

1.1 Worldwide Energy Consumption by Sector adapted from World

Resources Institute, 2003[23]. . . . . . . . . . . . . . . . . . . . . . 3
1.2 Annual growth of energy consumption from different sources during
1965 to 2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Energy demand relative to the size of their population & CO2
Intensity and Energy Consumption per GDP [22] . . . . . . . . . . 4
1.4 Average costs of fossil fuels for the electric power industry in the
United States from 2005 to 2023.[1] . . . . . . . . . . . . . . . . . . 5
1.5 Renewable energy demand growth by sector, main case, 2023-2030[20] 6
1.6 : Employment in the renewable energy sector in the year 2009 in all
EU-27 countries (EurObserv’ER 2010) . . . . . . . . . . . . . . . . 8
1.7 Tataouine the photovoltaic plant in Tunisia(capacity of 10 MW) |
Eni-2019 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.8 Chennai gets five new biogas plants to manage bio-waste . . . . . . 10
1.9 Modern renewable energy generation by source, World(Energy Insti-
tute - Statistical Review of World Energy (2024)) . . . . . . . . . . 11
1.10 Turbine Compenants (One Power Company) . . . . . . . . . . . . . 12
1.11 Types of wind turbines . . . . . . . . . . . . . . . . . . . . . . . . 13
1.12 Wind turbines power according wind velocity . . . . . . . . . . . . . 13
1.13 Wind Turbine Air Flow . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.14 Geometric features of airfoil. . . . . . . . . . . . . . . . . . . . . . . 18
1.15 Relationship between effective angle of attack, reduced frequency,
and Tip Speed Ratios. . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.16 Lift and drag force of wind turbine airfoil. . . . . . . . . . . . . . . 20

2.1 Workbench Ansys 19 Interface . . . . . . . . . . . . . . . . . . . . . 36

3.1 Darrieus wind turbine . . . . . . . . . . . . . . . . . . . . . . . . . 40

3.2 Savonius wind turbine D = 30 mm and D = 20 mm . . . . . . . . . 41
3.3 Helicoidal wind turbine . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.4 Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

6
3.5 Wind Tunnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.6 Electromagnetic brake (Deserti Meccanica 2025) . . . . . . . . . . . 44
3.7 Curve of Cp = f (λ) . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.8 Curve of Ct = f (λ) . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

4.1 3D simulation of the Savonius wind turbine in SolidWorks . . . . . 49

4.2 The two domains used in the simulation . . . . . . . . . . . . . . . 50
4.3 Domain mesh generated with 0.01 m element size . . . . . . . . . . 50
4.4 Inlet and outlet conditions . . . . . . . . . . . . . . . . . . . . . . . 51
4.5 Calculation residuals for flow around the helical Savonius (V = 4.7
m/s) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.6 Different simulation conditions realized in the same project . . . . . 53
4.7 Velocity distribution for V = 4.7 m/s and λ = 0.07 . . . . . . . . . 54
4.8 Turbulance distribution for V = 4.7 m/s and λ = 0.07 . . . . . . . . 54
4.9 Pressure around the blades for V = 4.7 m/s and λ = 0.07 . . . . . . 55
4.10 Curve of Cp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.11 Curve of Ct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.12 Curve of Cpexp and Cpnum . . . . . . . . . . . . . . . . . . . . . . 56
Introduction

The growing demand for sustainable energy solutions has intensified research
efforts in the field of renewable energy. As fossil fuel resources deplete and climate
change accelerates, the transition toward clean energy sources has become impera-
tive. Among these, wind energy represents one of the most promising alternatives
for large-scale power generation, offering an environmentally friendly and sustain-
able solution. In recent years, Computational Fluid Dynamics (CFD) has emerged
as a powerful tool for optimizing renewable energy technologies, particularly in
the domain of wind power. CFD allows for the detailed analysis of aerodynamic
performance, wake interactions, and turbulence effects in wind turbine designs. This
study focuses on the application of CFD in wind energy, specifically investigating
the performance of helical wind turbines, also known as vertical-axis wind turbines
(VAWTs).
This research integrates an experimental and numerical approach to compre-
hensively assess the aerodynamic characteristics and energy efficiency of a helical
wind turbine. The experimental phase involves constructing a scaled prototype and
conducting wind tunnel tests to collect data on power output, pressure distribution,
and turbulence effects. Simultaneously, advanced CFD simulations employing
high-fidelity turbulence models are performed to analyze airflow behavior under
varying wind conditions. By correlating experimental measurements with numerical
predictions, this study aims to enhance the accuracy of CFD models and improve
the design of wind turbines for maximum efficiency. The findings of this study will
contribute to the broader field of hydraulic and environmental engineering by pro-
viding valuable insights into the optimization of wind turbine performance through
CFD methodologies. Additionally, the results will aid in refining sustainable wind
energy solutions, ultimately supporting the global transition toward cleaner and
more efficient power generation systems.

1
Chapter 1

Overview Fundamental Concepts

1.1 Introduction
Renewable energy is essential for a sustainable future, with wind power playing
a key role due to its efficiency and low environmental impact. There are two main
types of wind turbines: Horizontal Axis Wind Turbines (HAWTs) and Vertical
Axis Wind Turbines (VAWTs), each claiming superiority over the other. While
HAWTs are the most commonly used, VAWTs can operate with wind from any
direction and require less space, making them a promising energy source for the
future.
This chapter covers energies and focuses on VAWTs, exploring their different
designs, operating principles, and aerodynamic aspects.

1.2 Energies
1.2.1 Energetic situation in the world
The annual consumption of various energy sources can be converted into Joules
and Watts to determine energy consumption per person in each country. Underde-
veloped countries, particularly in Africa, have the lowest rates, while developed
economies like North America, Australia, South Korea, Japan, and Europe have
consumption rates exceeding 3kW/person.

2
Figure 1.1: Worldwide Energy Consumption by Sector adapted from World Re-
sources Institute, 2003[23].

Living without services and in risky circumstances, they rely on solar energy,
natural processes, hunting, farming, and wood. Consumption rates in wealthy and
industrialized nations are 30-100 times higher than our bodies require, leading to
inefficient financial systems.
TThe 20th century saw a rise in energy sources like nuclear, hydroelectric, fossil
fuels, and renewables. Between 1965 and 2018, people burned 404 gigatonnes of
fossil fuels and emitted 1204 gigatonnes of CO2. Since 1980, fossil fuel consumption
has slightly declined.
Global CO2 emissions are expected to reach an all-time high, rising by over 2%
in 2018 due to increased fossil fuel use for economic purposes.[6]

Figure 1.2: Annual growth of energy consumption from different sources during
1965 to 2018.

1.2.2 Fossil energies

Since the world’s formation 4,5 billion years ago, our energy sources have
evolved, with the first living forms appearing 3 and a half billion years ago and more

3
sophisticated ones taking over 1.5 billion years ago. The main energy sources are
hydroelectricity (2.4%), nuclear power (2.5%), coal (25%), natural gas (25%), and
petroleum (45%). Fossil fuels now make up 95% of global energy consumption.[15]
In 2023, total energy consumption increased by 2% compared to 2022, with
renewable energy increasing six times faster. Natural gas demand decreased by 1%
, while electricity climbed by 2.5%. Coal production reached a record 179 EJ, oil
consumption surpassed one hundred million barrels per day, and CO2 emissions
increased by 2.1%, setting a new record.

Figure 1.3: Energy demand relative to the size of their population & CO2 Intensity
and Energy Consumption per GDP [22]

The post-COVID recovery saw strong growth in renewable energies, with 276
GW of installed capacity, including 77% solar and 33% wind, including 75% solar
additions by China. Wind power also increased, with 37 GW in China and 32 GW
in Europe. Biofuel production rose by 8%, with Brazil and the US dominating.
Nuclear power remained stable at 9%, while energy storage reached 55.7 GW, with
50% in China.[22]
The 2022 energy crisis increased costs, with natural gas prices in Europe reaching
over 200 USD/barrel, accounting for 90% of power generation costs. Maintenance,
CO2 prices, and capital costs contributed to the increase. Wholesale power prices
tripled in the first half of the year due to a 40% increase in electricity production
costs. LNG demand in Europe increased by 66% in eight months, but declined in
Asia since 2015.[2]
Depending on the source, fossil fuels in the US electric power sector have
different prices. In 2023, the average cost of fossil fuels was 3.12 USD per million
British thermal units (Btu), with coal costing 2.36 USD per million Btu and
petroleum costing 16.53 USD per million Btu.[1]

4
Figure 1.4: Average costs of fossil fuels for the electric power industry in the United
States from 2005 to 2023.[1]

1.3 Renewables energies

Renewable energy (RE) is energy from sources like biomass, geothermal heat,
wind, solar, hydropower, tides, waves, and ocean thermal energy that replenish
themselves at a rate equal to or higher than consumption, although overuse can
cause depletion.
Fossil fuels take millions of years to form, while renewable energy sources can
be replenished continuously. Non-renewable resources like nuclear power and fossil
fuels are the primary energy source for developed countries, influenced by nuclear
hazards and the 1970s energy crisis.
Renewable energy can provide a sustainable solution, but overuse of resources
like biomass and geothermal energy can lead to localized depletion.[17]
On The main-case forecast predicts that transportation, heating, and power
sectors will consume 60% more renewable energy between 2024 and 2030, with the
share of renewable energy in overall energy consumption increasing from 13% in
2023 to almost 20% by 2030, largely due to government support, declining costs,
and expanding electricity use. Renewable fuels including hydrogen, e-fuels, and
liquid, gaseous, and solid bioenergy account for almost 15% of the projected rise in
demand for renewable energy. enewable energy, mainly solar and wind, is expected
to account for 46% of electricity by 2030, meeting 20% of heat demand and 6%
of transportation needs. However, electricity will only account for 23% of overall
energy use.[20]

5
Figure 1.5: Renewable energy demand growth by sector, main case, 2023-2030[20]

1.3.1 Green Power energy

"Green power" refers to electricity produced using renewable resources that
offer the most environmental benefits. It can be optional or unregulated, and it
can surpass what is otherwise required. Renewable energy comes from natural
resources that replenish themselves over short periods without running out. Green
power refers to renewable energy sources with the most significant environmental
benefits.Citing :
*solar Low-impact
*hydropower resources
*wind
*geothermal energy,
*biogas
*biomass
and some types of plant and waste materials .Large hydropower and municipal
solid waste are commonly considered renewable energy sources, but they are often
not included in the concept of green power.[19]

1.3.2 Advantages of renewables energies

Green power offers financial, stakeholder, environmental, and domestic energy
benefits to users, enhancing purchasing organization relationships, promoting
stakeholder relations, and expanding domestic energy resources.

6
Environmental Benefits
Green power reduces carbon footprint, air pollution, and water impact by
adhering to the GHG Protocol Corporate Reporting Standard, reducing emissions
from fossil fuels, and minimizing thermal pollution compared to traditional power
plants.

Economic Benefits to Purchasing Organization

Technological advancements in wind and solar energy have reduced economic
costs, provided stable electricity prices through long-term contracts, and enhanced
energy security through on-site or third-party generation.

Stakeholder Relations
Green electricity helps businesses achieve environmental objectives, boost brand
credibility, demonstrate civic leadership, enhance public image, boost staff en-
gagement, attract students, and differentiate products and services. Participating
in environmental projects strengthens an organization’s reputation and attracts
investment. Companies that adopt green electricity also attract students and
provide a competitive advantage in sustainable supply chains.

Development of Domestic Energy Resources

Green power boosts the domestic economy by creating jobs in renewable en-
ergy manufacturing, installation, and operation. Costs depend on factors like
energy source, supplier, contract terms, purchase quantity, location, and incentives.
Cost-competitive options like self-generation and long-term contracts provide cost
stability. Renewable Energy Certificates (RECs) pricing fluctuates based on supply
and demand.

7
Figure 1.6: : Employment in the renewable energy sector in the year 2009 in all
EU-27 countries (EurObserv’ER 2010)

1.3.3 Projects to enhance the use of renewable energies

In Tunisia:
Current Renewable Energy Capacity , for June 2023, Tunisia’s installed renew-
able energy capacity totaled approximately 565 megawatts (MW), devided:

• Wind Power: 240 MW.

• Solar Power: 263 MW.

• Hydroelectric Power: 62 MW.

This combined capacity represents about 8% of the nation’s total energy production
capacity.[7]
-Key Projects and Initiatives
The Kairouan Solar Power Station, a 120 MW project, is set to be constructed
in late 2023 and operational by Q3 2025. [12]
Tunisia is collaborating with international developers to advance two 50 MW
solar power plants in Sidi Bouzid and Tozeur, as part of its strategy to expand
solar energy infrastructure.[10]

8
 The Sfax Waste-to-Methane Gas Project aims to convert solid waste into
methane gas through anaerobic digestion, producing 1.9 million cubic meters
annually to power 800 Sfax taxis. [11] -International Collaborations and Investments
Tunisia partners with H2 Global Energy and Amarenco to develop a $6 billion
green hydrogen and ammonia project, aiming to boost renewable energy, create
jobs, drive economic growth, and mitigate climate change.[9]
ACWA Power and Tunisian government signed a Memorandum of Understanding
in April 2024 to explore large-scale green hydrogen projects in Tunisia for export
markets and local consumption.[9]

In the world :
In 2024, clean energy sources like solar and wind power accounted for 92.5%
of new electricity capacity, with China leading the way with 64% of solar panel
installations. However, projections suggest a 28% failure to triple renewable energy
capacity by 2030. China’s 887 GW solar power surpasses the US and Europe,
underscoring the urgent need for global adoption.[8]
The United Nations Development Programme (UNDP), International Renewable
Energy Agency (IRENA), and Sustainable Energy for All (SEforALL) are leading
international efforts to promote sustainable energy worldwide. UNDP aims to
provide affordable, reliable energy to 500 million people by 2025, while SEforALL
focuses on ending energy poverty.
Expected Projects :
The Tengger Desert Solar Park in China, Noor Ouarzazate Solar Complex
in Morocco, and Neom Green Hydrogen Project in Saudi Arabia are examples
of solar installations that are integrating renewable energy. The Tengger Desert
Solar Park covers 43 square kilometers and uses over 20 photovoltaic power plants.
The Noor Ouarzazate Solar Complex uses mirrors to focus sunlight onto receivers,
producing steam to drive turbines and generate electricity. The Neom Green
Hydrogen Project is the world’s largest utility-scale hydrogen facility, integrating
3.9 gigawatts of renewable power from solar, wind, and storage. The European
Offshore Wind Deployment Centre aims to reduce offshore wind energy costs and
promote advancements. The Hornsdale Power Reserve showcases the potential of
battery storage in supporting renewable energy integration.

1.3.4 Types of renewables energies

Solar Energy : Solar energy, a clean, reliable resource, has the potential to
exceed human energy consumption. Governments support solar adoption through
subsidies, and a 50% cost reduction in solar panels could make it affordable. Solar

9
energy can be passive or active, with passive systems using sun’s thermal energy
for heating and active systems using semiconductor-based panels.[3]

Figure 1.7: Tataouine the photovoltaic plant in Tunisia(capacity of 10 MW) |

Eni-2019

Biomass Energy:Bioenergy, derived from agriculture and waste, is a renewable

energy source that can be produced through digestion, combustion, or decomposi-
tion. In Chennai, a government plant processes 40 tons of vegetable waste daily,
generating 4,800 electricity units at Rs. 3.15 per unit.[3]

Figure 1.8: Chennai gets five new biogas plants to manage bio-waste

Tidal Power :India’s coastal location offers significant potential for wave
energy, with areas averaging 65 MW per mile. The most cost-effective method is
through oscillating water columns, but wave power plants face challenges during
cyclones and severe storms, making them temporarily inoperative.[3]
Geothermal Energy: GGeothermal energy, derived from Earth’s heat, is a
clean and sustainable resource. It ranges from shallow ground to deep hot rock

10
and magma. Geothermal heat pumps use stable underground temperatures to heat
and cool buildings, extracting heat in winter and transferring excess in summer.[3]
Hydraulic energy : Utilizing the kinetic and potential energy of currents and
waterfalls, hydraulic energy—also known as water energy—allows us to generate
electricity by harnessing the movement of water.
Green Hydrogen:Green hydrogen is a clean, sustainable energy source that
generates no carbon emissions, generating hydrogen and oxygen from water using
electrolysis powered by renewable sources like wind or solar.[14]
Wind Power:Wind energy is generated by harnessing atmospheric air move-
ment, driven by the sun’s heat. Wind turbines, placed on 30m towers, convert
wind into mechanical power, producing 50-350 kW. They operate at 22 km/h and
are clean, renewable, and free from pollution or hazardous waste.[3]

Figure 1.9: Modern renewable energy generation by source, World(Energy Institute

- Statistical Review of World Energy (2024))

Opening up to integration in hybrid systems: example: photovoltaic-

powered agricultural pumping systems : Photovoltaic pumping is popular in
agricultural applications, involving a motor-driven pump, solar panels, charge
controller, and storage batteries. Battery-powered pumping avoids adaptation
issues and ensures consistent pumping throughout the day.[24]

1.4 Wind Power

Over the past 15 years, the global demand for wind turbines has significantly
increased, with an estimated 24,000 MW of wind energy capacity currently operating.
This demand is driven by the growing demand for electric power plants using cleaner
fuels. Windfarms with multiple turbines are being built, and the average size of a
single wind turbine has grown from 100 kW to 1 MW or more. Large-scale wind
energy projects are now producing electricity at prices comparable to traditional
power plants.[23]

11
1.4.1 Main components of a wind turbine:
Wind turbines come in various configurations, but a typical model consists of
three main components:
The Tower: Its height depends on the turbine’s power and is usually made of
metal for durability. It supports key elements like the nacelle and rotor. Towers
can reach up to 100 meters to optimize wind capture and allow the use of longer
blades.
The Nacelle: This essential part houses the generator, which converts kinetic
energy into electricity. It also contains control mechanisms, such as brakes and
wind-tracking systems, ensuring safe operation by adjusting blade orientation and
stopping the turbine in extreme conditions.
The Rotor: Comprising a hub and three blades positioned at the front of
the nacelle, it generates mechanical energy from wind movement. This energy is
then converted into electricity and transmitted via cables inside the tower to a
distribution unit.

Figure 1.10: Turbine Compenants (One Power Company)

1.4.2 Types of Wind Turbines:

Wind turbines are classified based on their axis orientation and operational
characteristics. The main types include:
1. Horizontal-Axis Wind Turbines (HAWTs) Common upwind turbines
rotate blades horizontally and are mounted on tall towers to capture stronger winds.
2. Vertical-Axis Wind Turbines (VAWTs) Blades rotate around a vertical

12
axis, capturing wind from any direction.
3. Specialized Wind Turbines Offshore wind turbines, floating turbines, and
hybrid turbines .

Figure 1.11: Types of wind turbines

=> In our case of study , we will be conducting experience and simulation on a

Savonius VAWT.

1.4.3 HAWT and VAWT

Compared to the VAWT, which can only reach 40% efficiency and is best suited
for small capacities at city sites and offshore areas, the HAWT is better appropriate
for big capacities in distant and offshore locations. Its efficiency also reaches 50%.
Compared to the VAWT, the HAWT has a 2% greater power coefficient. However,
because HAWTs are higher above the ground, they are typically more costly than
VAWTs and require more maintenance. It is impossible to give up either kind. In
contrast to the VAWT, which can only reach 6 Mw of capacity with active pitch
control, the current HAWT can reach 20 Mw or more. Valley, mountain, sea, and
land breezes are among the occurrences that can be used to create power using
wind energy, as are times when they should be taken advantage of.[18]

Figure 1.12: Wind turbines power according wind velocity

[18]

13
1.4.4 Total efficiency of a wind turbine :
• 59.3% is the highest efficiency with which wind energy can be extracted and
transformed into electrical energy. The Betz limit is this.
• Current wind turbines are capable of exceeding this limit by 80%.
• Nevertheless, the wind fluctuates.
• Wind variability, the Betz limit, and turbine efficiency are combined to produce
the "Load Factor."
• Turbine capacity are rated at maximum output in winds of about 12 m/s.
• 30% to 35% are good onshore load factors.
• 35% to 40% are good offshore load factors.(OSCE)

1.4.5 Choosing a wind turbine for a site in Tunisia

The site analysis involves identifying the region, analyzing wind data, and
considering local constraints such as the power grid, protected areas, and dwellings.
The technical choice of wind turbine depends on the turbine size, mast height,
and power curve. The energy and economic analysis includes calculating annual
production, calculating the load factor, and analyzing the purchase and installation
costs. Environmental and social impact studies are conducted to assess noise,
wildlife impact, and local permits and authorizations. The regulations in Tunisia
include obtaining STEG approval and production permits for large-scale installa-
tions, complying with the Electricity Code, and environmental standards. The final
selection and installation are based on compatibility with the wind profile, energy
yield, cost and return on investment, reliability and availability of parts/equipment
in Tunisia, site planning, installation + grid connection, and commissioning and
monitoring.
Example of wind turbines adapted to Tunisia

Model Power Cut-in Speed Use Case

Bergey Excel 10 10 kW 2.5 m/s Homes, farms
XANT M-21 100 kW 3 m/s Off-grid, small business
Vestas V52 850 kW 3.5 m/s Rural, medium projects
GE 1.5sle 1.5 MW 3.5 m/s Wind farms

Table 1.1: Wind Turbine Examples (Short List)

Best locations in Tunisia

14
 Region Avg Wind Speed @100m Wind Potential
Cap Blanc (Bizerte) 7.5 – 8.0 m/s Very High
Sidi Daoud (Nabeul) 7.0 – 7.5 m/s Very Good
El Haouaria ∼7.0 m/s Very Good
Kef (Northwest) 6.5 – 7.0 m/s Good
Tataouine 6.5 – 7.0 m/s Good
Gabès 6.0 – 6.5 m/s Moderate to Good
Kasserine 5.5 – 6.0 m/s Moderate

Table 1.2: Most favorable wind sites in Tunisia(Global Wind Atlas)

1.5 Aerodynamic Concepts

1.5.1 BETZ Theory
According to the Betz Limit, also known as the Betz Law, which was determined
by German physicist Albert Betz about a century ago, no wind turbine generator
can transform more than around 60% of the wind’s kinetic energy into mechanical
(or electrical) energy by merely rotating the rotor blades.In general, the power (P)
produced by a wind turbine generator (WTG) is proportional to the volumetric
flow of wind across it. As energy is taken, this causes a pressure differential between
the wind entering and exiting the swept area of the rotor blades.
The power content in a cylindrical column of free unobstructed air moving at a
constant speed V is the rate of change in its kinetic energy:
dE
P = (1.5.1)
dt
Expressing the kinetic energy as:
1
E = mV 2 (1.5.2)
2
Substituting Eq. (2) into Eq. (1), we get:
d 1
 
P = mV 2 (1.5.3)
dt 2
Applying the chain rule of differentiation:
d dv du
(uv) = u + v (1.5.4)
dt dt dt
15
Then:

1
!
dV dm
P = 2mV +V2 (1.5.5)
2 dt dt
For a constant wind speed V :
dV
=0 (1.5.6)
dt
1d
P = (mV 2 ) (1.5.7)
2 dt
And consequently, the power can be expressed in terms of its speed V and its mass
flow rate as:
1
P = ṁV 2 (1.5.8)
2
If the cross-sectional area of the column of air is S, and its density is ρ, the mass
flow rate is:
ṁ = ρSV (1.5.9)
By substituting from Eq. (4) into Eq. (3), this yields the power content of the
column of air as:
1
P = ρSV 3 (1.5.10)
2
If the diameter of the column of air is D, then:

1 πD2 3
P = ρ V (Ragheb2021) (1.5.11)
2 4
*The Law of Mass Conservation

S0 V0 = SV = S2 V2 (1.5.12)

*Power Absorbed by the Rotor

P = SpV 2 (V0 − V2 ) (1.5.13)

*Kinetic Energy Variation

1
∆EC = SpV (V02 − V22 ) (1.5.14)
2
*Bernoulli’s Equation
1 V0
V = (V0 + V2 ) and V2 = (1.5.15)
2 3

16
 Figure 1.13: Wind Turbine Air Flow

*Final Expression
2 2 16 1
 2    
PMAX = Sp V0 V0 = SpV03 (1.5.16)
3 3 27 2
Thus, the theoretical maximum power developed by the wind turbine is equal to
16
27
of the power of the incident wind passing through the wind turbine. Therefore,
the Betz coefficient is given by:
16
= 0.592593 = 59.26%
27

1.5.2 Aerodynamic variables:

Aerodynamics and Airfoil Geometry
Aerodynamics is a branch of fluid dynamics that studies the motion of air
around a body. This body may either move through a stationary fluid or remain
fixed while the fluid moves around it (Paraschivoiu, 1998).
An airfoil refers to the cross-sectional shape of a wing, blade (propeller, rotor,
or turbine blade), or any aerodynamic surface. The design of airfoils is a crucial
aspect of aerodynamics, as it directly impacts performance and efficiency. There is
a wide variety of airfoil shapes, each tailored to specific functional requirements
and manufacturing constraints.
The geometry of an airfoil is defined by several key parameters:
Leading edge: The foremost point of the airfoil.
Trailing edge: The rearmost point of the airfoil.
Upper surface (extrados): The top part of the airfoil.
Lower surface (intrados): The bottom part of the airfoil.
Chord line: A straight line connecting the leading edge to the trailing edge.
Chord length (c): The length of the chord line, serving as the reference
dimension for the airfoil cross-section.

17
 Figure 1.14: Geometric features of airfoil.
[4]

Camber line: The curve representing the midpoint between the upper and
lower surfaces. Its shape depends on the distribution of thickness along the chord.
Maximum thickness: The greatest distance between the upper and lower
surfaces. The relative thickness (expressed as a percentage) is calculated as the
ratio of the maximum thickness to the chord length (Radi, 2020).
*The relative velocity of the airflow over the profile is the vectorial sum of the
disturbance wind (from airfoil motion) and airfoil displacement speed. If the wing is
stationary, the relative velocity is simply the undisturbed wind speed (Fady,2011).
Q
V = (1.5.17)
A
Where: V:Air velocity (m/s) Q:AirFlow rate (m3) A:Sectional Area (m2)
Solidity
sigma
is defined as the ratio between the total blade area and the projected area of the
turbine (Carrigan, 2011). This term is expressed as:
c
σ=N (1.5.18)
2R
Where: N: Number of blades C: Chord length (m) R: Rotor radius (m)
This formula does not apply to HAWTs (Horizontal Axis Wind Turbines) as
they have a different swept area shape (Carrigan, 2011).
*Angle of attack: As the rotor rotates, the local azimuth of each blade
changes and the W and alpha values change. The given angle of attack can be
expressed as follows:

sin θ
!
αθ = tan −1
, (1.5.19)
λθ + cos θ

18
Alpha is influenced by the specific velocity ratio, with a small ratio causing a high
angle of attack and a high ratio causing a low angle. This value is crucial when
large variations in wind turbine speed lead to aerodynamic stalls and can increase
the cyclic load of the turbine. Therefore, the specific speed ratio plays a significant
role in determining alpha’s behavior.(Brian,2012).

Figure 1.15: Relationship between effective angle of attack, reduced frequency, and
Tip Speed Ratios.
[4]

1.5.3 Aerodynamic forces

Lift Force (L): A low angle of attack on an airfoil allows airflow to remain
attached to the surface, creating a low-pressure zone on the extrados and a high-
pressure zone on the intrados. This pressure difference generates lift, a force
perpendicular to the wind direction, crucial for wind turbine rotation. Drag
Force (D): Drag, a component of airflow, is opposite to fluid flow and aligns with
air velocity. It consists of friction drag and pressure drag, influenced by surface
roughness and airfoil shape.
Attached flows primarily generate friction-based drag, with lift dominating.
High angles of attack can lead to airflow separation, causing low-pressure wake and
stalling.[16]

19
 Figure 1.16: Lift and drag force of wind turbine airfoil.

1.5.4 Aerodynamic coefficients

Reynolds number Re The Reynolds number is an adimensional quantity
used in fluid mechanics to characterize the flow regime of a fluid around a moving
object. It is defined as the ratio between the fluid’s inertial and viscous forces
(Canonne, 2000).
VD ρV D
Re = = (1.5.20)
ν µ
Lift (CL) and drag (CD), two dimensional coefficients, are the primary determinants
of aerodynamic forces like lift and drag. These coefficients are independent of the
profile’s size but particular to the aerodynamic profile’s shape (Jamati, 2011).

D
CD = 1 (1.5.21)
2
ρA(V − v)2

L
CL = 1 (1.5.22)
2
ρA(V − v)2
With q : dynamic pressure : V: undisturbed wind speed. rho: air density D :
Diameter For a blade section (2D): S=c profile chord. For a rotor: A = swept area.
Coefficient of power, Cp a dimensionless metric called the power coefficient
gauges a wind turbine’s aerodynamic efficiency. According to Jamit (2011), it is
the ratio of the power generated by the rotor to the power available in the wind.
1
q = ρV 2 (1.5.23)
2
P ωC
Cp = = 1 (1.5.24)
Pmax 2
ρAV 3

20
where A is the cross-sectional area swept by the wind, C is the rotor torque, and

is the rotor’s angular speed. Cp < 1. According to Armano (2014), Darrieus

machines often have a much greater power coefficient than Savonius machines.
0.593 is the highest value that corresponds to the Betz limit. The coefficient of
speed a dimensionless metric known as specific speed, or TSR (tip speed ratio),
creates a connection between the upstream flow velocity and the speed at the wind
turbine blade tips. The rotor’s radius and angular speed

determine this parameter (Armano, 2014). The following defines its expression:

R
λ=ω (1.5.25)
v
coefficient of torque This coefficient represents the mechanical performance of
the wind turbine and is used to evaluate the torque value at various operating
positions. It is described using the following formula as :
Cp
Ct = (1.5.26)
λ

1.6 Dimensional Analysis

Dimensional analysis is a crucial method in engineering and physics to simplify
systems by identifying fundamental parameters. In wind turbine design, it helps
researchers understand relationships between variables like geometry, fluid dynamics,
and operational conditions, enabling scaling analysis and performance prediction
for various turbine designs and wind conditions.

1.7 Key Parameters in the Savonius Wind Tur-

bine
The performance of a Savonius wind turbine, especially its helical configuration,
is influenced by key parameters such as turbine geometry, wind properties, and
operational conditions such as:

21
1. Geometric Parameters Rotor Diameter (D): Rotor diameter, including
blades, significantly influences turbine swept area, affecting wind energy capture.
A larger diameter increases swept area, enhancing turbine efficiency.
Blade Height (H): The height of turbine blades significantly impacts air
swept volume, blade-wind interaction, and turbine energy extraction capacity.
Blade Overlap (O): The Savonius turbine’s blade overlap ratio, ranging from
0.25 to 0.5, influences torque generation and efficiency, with too large an overlap
increasing drag.
Helical Twist Angle (θ): Helical Savonius turbines twist blades, affecting
aerodynamic performance by modifying wind attack angle, improving efficiency,
reducing drag, and enhancing self-starting capability.
Number of Blades (N): Savonius turbines have two or three blades, affecting
torque generation and rotation smoothness. More blades increase torque but
increase drag, while fewer reduce drag but require higher wind speeds.

2. Fluid Properties Air Density (ρ): Air density, influenced by altitude,

temperature, and humidity, affects wind energy availability. Higher density leads
to increased wind energy, enhancing turbine performance:
p
ρ= (1.7.1)
Rspécifique T
Air Viscosity (µ): The dynamic viscosity of air influences wind frictional
resistance, affecting turbine blade formation and turbine efficiency in converting
wind energy into mechanical power:
3/2
a T
 
µ = µ0 (1.7.2)
b T0
Wind Speed (V): Wind speed significantly impacts wind turbine perfor-
mance, directly affecting the kinetic energy available in the wind, which increases
exponentially as the speed increases:
1
P = ρAv 3 (1.7.3)
2

3. Operational Parameters Rotational Speed (ω): The angular velocity

refers to how fast the turbine blades rotate. It affects the tip speed ratio (TSR), a
key factor in determining the turbine’s efficiency:

V = 2πrω (1.7.4)

22
 Torque (T): Torque measures turbine’s wind energy conversion into rotational
motion, influenced by blade aerodynamics and wind speed:
P
τ= (1.7.5)
ω
where:

• T = Torque (N·m)

• P = Power output (W)

• ω = Angular velocity (rad/s)

Power (P): The power output of a turbine is a function of the wind speed,
the swept area, and the turbine’s efficiency in converting wind energy:

P = Fν (1.7.6)
Power Coefficient (Cp ): The power coefficient, a dimensionless parameter,
measures a turbine’s efficiency in converting wind kinetic energy into mechanical
energy, with a theoretical maximum of 0.593 (Betz limit):

P
Cp = (1.7.7)
0.5ρAV 3
where:

• P = Power output (W)

• ρ = Air density (kg/m3)

• A = Swept area of the turbine (m2)

• V = Wind speed (m/s)

4. Other Important Parameters Tip Speed Ratio (TSR): The tip speed
ratio (TSR) is crucial for determining turbine efficiency:
ωD
λ= (1.7.8)
2V
where:

• λ = Tip speed ratio (dimensionless)

• ω = Angular velocity (rad/s)

23
 • D = Rotor diameter (m)

• V = Wind speed (m/s)

Reynolds Number (Re): The Reynolds number reveals turbulent flow around
blades:
ρV D
Re = (1.7.9)
µ
where:

• Re = Reynolds number (dimensionless)

• ρ = Air density (kg/m3)

• V = Wind speed (m/s)

• D = Rotor diameter (m)

• µ = Dynamic viscosity of air (Pa·s)

1.8 Buckingham π Theorem

If there are n variables in a problem containing m primary dimensions, the
equation relating all variables will have (n − m) dimensionless groups:

πi = f (π2 , π3 , . . . , πn−m )
The π groups must be independent of each other [25].

1.8.1 Key Dimensionless Groups

For our isothermal, incompressible fluid study, dimensionless parameters can
be written as the product of powers of three fundamental quantities (mass, time,
length):

1.9 Conclusion
Chapter 1 provided a literature review on wind energy applications, discussing
the advantages and limitations of various VAWT systems. Building on this founda-
tion, this chapter explored different VAWT designs, their operating principles, and
key aerodynamic aspects to assess their performance. VAWTs present a promising

24
 Table 1.3: Dimensionless groups related to wind turbine analysis

Π Groups Description

Π1 = 1/Re Inverse Reynolds number

Π 2 = Cp Power coefficient
Π3 = λ = ω/RV Tip speed ratio
Π4 = CT = Cp /λ Torque coefficient
Π5 = ARblade = l/c Blade aspect ratio
Π6 = ARrotor = l/2R Rotor aspect ratio

alternative to HAWTs, offering the ability to capture wind from any direction while
requiring less space. Understanding their aerodynamics is essential for optimizing
efficiency and functionality. As the demand for renewable energy continues to grow,
further advancements in VAWT technology could enhance their role in sustainable
power generation.The next chapter includes an experimental study of a Savonius
VAWT using those concepts .

25
Chapter 2

Software

2.1 Introduction
In this chapter we will introduce the software , its tools ,equations and models
used to simulate different conditions .

2.2 CFD
2.2.1 CFD
What is CFD? CFD stands for Computational Fluid Dynamics.
Computational fluid dynamics (CFD) is the study of predicting fluid properties
like heat transfer, mass transfer, and chemical reactions by solving differential
equations from numerical databases.
Fluid flow is typically studied using two methods: fluid dynamics in experimentation
and fluid dynamics theory.
Why use CFD? Advantages of using CFD over experimental methods: -
CFD offers a high-fidelity database for diagnosing flow fields and simulating physical
phenomena like nature, environmental effects, and hazardous phenomena, making
them difficult to experiment with.
In practice, CFD is used for:
- Design studies and optimization of (new) products;
- Performance studies of systems/installations;
- Providing complementary data to theoretical and experimental data.[13]
• CFD Applications
Some examples of applications include: Aerodynamics , Hydrodynamics , Renewable
energies , Turbomachinery , Electronics,Chemical process engineering ,Marine

26
 Experimental Measurement CFD Simulation

Expensive Less expensive

Slow Fast
Sequential Parallel
Single-use Multi-use

Table 2.1: Comparison between Experimental Measurement and CFD Simulation

engineering ,Combustion , Environmental engineering ,Biomedical engineering ,etc

2.2.2 General equations of incompressible flows

The general equations of incompressible fluid flow are the conservation of mass
and momentum equations, also known as the Navier-Stokes equations. These
equations are as follows: U is the fluid velocity Einstein’s convention :
∂ui
= 0, i = 1, 2, 3 (2.2.1)
∂xi
ρ is the density of the fluid, t is the time , p is the pressure, µ is the dynamic
viscosity of the fluid, f is the external force. Continuity equation (Mass conservation
equation):

⃗ + Dρ = 0
ρ div U (2.2.2)
Dt
Conservation equation of momentum in vector notation :

∂ui ∂ui 1 ∂p ∂ 2 ui
+ uj =− +ν + fi (2.2.3)
∂t ∂xj ρ ∂xi ∂xj ∂xj

Using the index form the previous equation becomes :

∂ui ∂ui 1 ∂p ∂ 2 ui
+ uj =− +ν + fi (2.2.4)
∂t ∂xj ρ ∂xi ∂xj ∂xj

(cours Chahed, Turbulance,2024)

27
Averaged Navier-Stokes equations
The general form of the incompressible Navier-Stokes equation for a fluid is:
!
∂⃗v
ρ + ⃗v · ∇⃗v = −∇p + µ∇2⃗v + f⃗
∂t
where:

• ρ: Fluid density (kg/m3)

• ⃗v : Velocity vector field (m/s)

• p: Pressure (Pa)

• µ: Dynamic viscosity (Pa·s)

• f⃗: Body forces (e.g., gravity)

Reynolds Decomposition
To model turbulent flows, we decompose the instantaneous variables into their
mean and fluctuating parts:

⃗v = ⃗v¯ + ⃗v ′
p = p̄ + p′
where:

• ⃗v¯: Time-averaged velocity

• ⃗v ′ : Fluctuating velocity

• p̄: Time-averaged pressure

• p′ : Fluctuating pressure

Substituting into the Navier-Stokes Equation

Substituting the Reynolds decomposition into the original Navier-Stokes equa-
tion, we get:

∂(⃗v¯ + ⃗v ′ )
!
ρ + (⃗v¯ + ⃗v ′ ) · ∇(⃗v¯ + ⃗v ′ ) = −∇(p̄ + p′ ) + µ∇2 (⃗v¯ + ⃗v ′ ) + f⃗
∂t

28
Simplifying the Terms
Time Derivative
The time derivative of ⃗v becomes:

∂(⃗v¯ + ⃗v ′ ) ∂⃗v¯ ∂⃗v ′

= +
∂t ∂t ∂t
Since the fluctuating velocity averages to zero, we get:

∂(⃗v¯ + ⃗v ′ ) ∂⃗v¯
=
∂t ∂t

Convective Term
The convective term ⃗v · ∇⃗v expands as:

(⃗v¯ + ⃗v ′ ) · ∇(⃗v¯ + ⃗v ′ ) = ⃗v¯ · ∇⃗v¯ + ⃗v¯ · ∇⃗v ′ + ⃗v ′ · ∇⃗v¯ + ⃗v ′ · ∇⃗v ′

Since the time average of the fluctuating part ⟨⃗v ′ ⟩ = 0, we obtain:

(⃗v¯ + ⃗v ′ ) · ∇(⃗v¯ + ⃗v ′ ) = ⃗v¯ · ∇⃗v¯ + ⃗v ′ · ∇⃗v ′

Pressure Gradient Term

The pressure term becomes:

−∇(p̄ + p′ ) = −∇p̄ − ∇p′

Since ⟨p′ ⟩ = 0, the fluctuating pressure p′ does not contribute to the mean equation.

Viscous Term
The viscous term becomes:

µ∇2 (⃗v¯ + ⃗v ′ ) = µ∇2⃗v¯ + µ∇2⃗v ′

Final Averaged Navier-Stokes Equation

After simplifying, the **Averaged Navier-Stokes equation** becomes:

∂⃗v¯ ¯
!
ρ + ⃗v · ∇⃗v¯ = −∇p̄ + µ∇2⃗v¯ + f⃗ − ∇ · (⃗v ′ ⊗ ⃗v ′ )
∂t
Where the last term, ∇ · (⃗v ′ ⊗ ⃗v ′ ), represents the Reynolds stress tensor, which
accounts for the turbulent flux of momentum.(cours Chahed, Turbulance,2024)

29
Reynolds Stress Transport Equation
The transport equation for the Reynolds stress tensor component u′i u′j is:

∂u′i u′j ∂u′ u′

+ uk i j = Pij + Φij − εij + Dij
∂t ∂xk
where:
 
∂uj
• Pij = − u′i u′k ∂xk
+ u′j u′k ∂x
∂ui
k
is the production term,

• Φij is the pressure-strain correlation,

∂u′ ∂u′
• εij = 2ν ∂xki ∂xkj is the dissipation term,

• Dij includes molecular and turbulent diffusion:

∂u′i u′j ∂u′i u′j u′k 1 ∂p′ u′i ∂p′ u′j

! !
∂
Dij = ν − + +
∂xk ∂xk ∂xk ρ ∂xj ∂xi

(cours Chahed, Turbulance,2024)

Production Term Pij

The production term represents the generation of turbulence due to velocity
gradients and is given by:
∂vk
Pij = −vi′ vk′
∂xj
This term shows how turbulent kinetic energy is generated from the mean
velocity gradients.

Dissipation Term ϵij

The dissipation term represents the energy conversion from turbulence to heat
due to viscosity:
! !
∂vi ∂vj
ϵij = ν
∂xk ∂xk
This term accounts for the energy loss in turbulence due to viscosity.

30
Diffusion Term Φij
The diffusion term accounts for the molecular diffusion of Reynolds stresses
and is expressed as:
!
∂τij
Φij = νt
∂xk
where νt is the turbulent viscosity.

Reynolds Stress Tensor

represents the turbulent flux of momentum in a fluid and is a crucial component
in turbulence modeling. defined as:

τij = vi′ vj′

where:
• vi′ and vj′ are the fluctuating components of the velocity in the i-th and j-th
directions, respectively.
• vi′ vj′ represents the time-averaged product of the fluctuating velocity compo-
nents.
The Reynolds stress tensor τij is a that quantifies the momentum transfer in the
turbulent flow due to velocity fluctuations.

Components of the Reynolds Stress Tensor

For a three-dimensional flow, the Reynolds stress tensor has six independent
components:  
τ11 τ12 τ13 
 
τij = τ21 τ22 τ23 
 
 
 
τ31 τ32 τ33
where:
• τ11 is the Reynolds stress in the x-direction due to fluctuations in the x-
component of velocity.
• τ12 = τ21 is the Reynolds stress due to the interaction of velocity fluctuations
in the x-direction and the y-direction.
• τ33 represents the Reynolds stress in the z-direction.
(cours Chahed, Turbulance,2024)

31
2.2.3 Turbulance Models
Turbulence models can be classified into 3 models:
- 0-equation model (Prandtl).
- 1-equation model (Spalart-Allmaras).
- 2-equation model (k -
ϵ
,k-
ω
,SST ... )
Turbulence models are used to close the system of equations, as directly solving
the Reynolds stress transport equation is complex. One widely used model is the
k-eplison turbulence model.

k-epsilon Turbulence Model

The **k- epsilon model** is a two-equation model that solves for the **turbulent
kinetic energy (k)** and the **dissipation rate (epsilon)**.
!
∂k ∂k ∂ ∂k
+ vi = Pk − ϵ + νt
∂t ∂xi ∂xi ∂xi

Where: - Pk is the production term for turbulence kinetic energy k, - νt is the

turbulent viscosity, and - C1 , C2 are empirical constants.
This equation is used to solve for the turbulent kinetic energy and its dissipation
rate, which in turn can be used to compute the Reynolds stresses.

k-ω Turbulence Model

The k-ω turbulence model is a widely used two-equation model for simulating
turbulent flows. It involves solving transport equations for:

• k: turbulent kinetic energy,

• ω: specific dissipation rate (related to the dissipation rate ε by ω ≈ ε/k).

Transport Equation for k

" #
∂k ∂k ∂ ∂k
+ Uj = Pk − β ∗ kω + (ν + σk νt )
∂t ∂xj ∂xj ∂xj

32
Transport Equation for ω
" #
∂ω ∂ω ω ∂ ∂ω
+ Uj = α Pk − βω 2 + (ν + σω νt )
∂t ∂xj k ∂xj ∂xj

Turbulent Viscosity

k
νt =
ω

Production Term
!
∂Ui ∂Uj ∂Ui
P k = νt +
∂xj ∂xi ∂xj

Model Constants
Typical values for the standard k-ω model constants are:

α = 5/9,
β = 0.075,
β ∗ = 0.09,
σk = 2.0,
σω = 2.0.

• Accurate near-wall behavior without the need for wall functions.

• Suitable for internal flows and flows with strong adverse pressure gradients.

• Less accurate in free shear flows compared to k-ε.

SST k-ω Model

The Shear Stress Transport (SST) k-ω model combines the robustness of the
k-ω model near the wall with the free-stream independence of the k-ε model in the
outer flow region.

33
The k-l Turbulence Model
The k-l turbulence model is a two-equation model used to simulate turbulent
flows by solving for:
• k: turbulent kinetic energy,
• l: turbulence length scale.
The viscosity is defined as:
νt = Cµ k 1/2 l
Transport Equation for k
" #
∂k ∂k ∂ ∂k
+ Uj = Pk − ε + (ν + σk νt )
∂t ∂xj ∂xj ∂xj
Transport Equation for l
" #
∂l ∂l l ∂ ∂l
+ Uj = C1 lS + C2 Pk + (ν + σl νt )
∂t ∂xj k ∂xj ∂xj
Auxiliary Relations
3/2
• Turbulent dissipation rate: ε = Cε k l
• Production of turbulence:
!
∂Ui ∂Uj ∂Ui
Pk = νt +
∂xj ∂xi ∂xj
Model Constants
Cµ = 0.09,
Cε = 0.1643,
C1 = 0.33,
C2 = 0.57,
σk = 1.0,
σl = 1.0.

Applications
The k-l model is well-suited for:
• Low-Reynolds-number and transitional flows,
• Compressible boundary layers,
• Aerospace applications.

34
2.2.4 Numerical Simulation
RANS
The review discusses the Reynolds-averaged Navier-Stokes equations (RANS)
method for modeling turbulent flows, focusing on the limit of incompressible
flows with constant characteristics. It presents various RANS turbulence models,
including zero-equation, one-equation, two-equation, stress-equation, and algebraic-
stress models. The most popular modeling approaches and closures are described,
along with the unstable RANS methodology and a portion devoted to hybrid
RANS/large approaches.[21]

DNS
Direct Numerical Simulation (DNS) is the most accurate method for simulating
turbulent flows. It involves solving the full Navier-Stokes equations without any
turbulence models, thereby directly resolving all turbulent scales, from the smallest
dissipative scales to the largest energy-containing eddies.

Key Features
• Exact Solution

• No Turbulence Modeling

• High Computational Cost

• Ideal for Research

LES
Large-scale simulation (LES) primarily simulates flow energy on vast scales,
filtering and modeling turbulence scales below mesh size. However, this method
requires fine discretization due to small turbulence scales in the thin boundary
layer.

ANSYS 19
Using sophisticated solver options such as linear dynamics, nonlinearities, ther-
mal analysis, materials, composites, hydrodynamic, explicit, and more, Ansys
Mechanical is a finite element analysis (FEA) program used for structural anal-
ysis. ANSYS 19 provides increased productivity and efficiency in a number of
fields.Ansys Mechanical’s capabilities in acoustics, fracture, nonlinear capabilities,

35
FSI, performance, and general software usability were all improved with the release
of Ansys 19.(2024 ANSYS, Inc)

Ansys Workbench
ANSYS Workbench19 is the main GUI platform that links all simulation tools
like geometry creation, meshing, solvers (Fluent, Mechanical), and post-processing
in one place. It provides a visual, drag-and-drop workflow to easily set up and
manage simulations.(ANSYS Inc., 2019)

2.2.5 Modeling Turbulent Flow with ANSYS workbench

in 5 Steps including CFX Project
• Import or Create Geometry:Use DesignModeler or SpaceClaim to build
or import geometry.

• Generate Mesh:Open the Meshing module to define and create the compu-
tational grid.

• Set Up Physics:Define boundary conditions, materials, and solver settings

(e.g., in Fluent or Mechanical).

• Run Simulation (Solve): Start the simulation to compute results.

• Post-Processing(results,curves,profiles):Use tools like CFD-Post or Mechan-

ical’s result viewer to analyze outputs.

• Parameter Study (Optional)

Figure 2.1: Workbench Ansys 19 Interface

36
2.2.6 Finite Elements Method
TNumerous engineering disciplines, including fluid mechanics, heat and mass
transfer, and petroleum engineering, have made extensive use of the finite volume
method, a discretization technique that is well suited for the numerical simulation of
different types of conservation laws (elliptic, parabolic, or hyperbolic, for example).
The finite volume approach and the finite element method share certain key
characteristics, including the ability to be applied to various geometries, the use of
structured or unstructured meshes, and the production of robust schemes.[5]

2.3 Conclusion
This chapter is an initiation to the CFD with its components and elements used
to simulate different flows.

37
Chapter 3

Experimental Study of a
Vertical-Axis Wind Turbine
(Savonius Wind Turbine)

3.1 Introduction
This chapter treats the experimental study of the prototype of the Savonuis
VAWT and presents the results and interpretations

3.2 Geometric Features of the Helical Savonius

The rotor consists of two blades mounted on a shaft with end plates. Key
geometric parameters include:

• D: Rotor diameter (300 mm)

• H: Rotor height (194 mm)

• De : End plate diameter (191 mm)

• A: Swept area (0.0582 m2 )

• N : Number of blades (2)

38
 Parameter Value
D 300 mm
H 194 mm
De 191 mm
A 0.0582 m2
N 2

Table 3.1: Geometric parameters of the scale model

The full-scale prototype parameters were scaled by a factor of 10:

Parameter Value
D 3.0 m
H 1.94 m
De 1.91 m
A 3.88 m2
N 2

Table 3.2: Geometric parameters of the full-scale prototype

The overlap ratio, also known as the Overlap Ratio, defined as the ratio between
the overlap distance (e) and the chord length (d) of a wind turbine blade, plays a
crucial role in improving the rotor’s power coefficient.
e
OR = (3.2.1)
d
A scale model that is ten times the size of the actual prototype is what we
have suggested. The geometrical specifications of the helical Savonius of the actual
model are displayed in the table below.

3.3 Wind Turbines Available in our Laboratory

Darrieus wind turbine
One kind of vertical-axis wind turbine (VAWT) is the Darrieus, which is
distinguished by its unusually curved, eggbeater-like shape. It was created in the

39
 Parameters Values
D 3000 mm = 3 m
H 1940 mm = 1.94 m
De 1910 mm = 1.91 m
A 3.88 m2
N 2

Table 3.3: Geometric parameters of the real “Savonius Helicoidal” model

1930s by French engineer Georges Darrieus and works by using the aerodynamic
lift that the wind creates on the blades. The Darrieus turbine is ideal for urban or
turbulent wind conditions because, in contrast to horizontal-axis turbines, it can
capture wind from any direction. However, because of its low starting torque, it
usually needs an external power source or auxiliary mechanism to start spinning.
Despite this, it is an intriguing choice for the production of renewable energy due
to its straightforward design and capacity for small installations.

Figure 3.1: Darrieus wind turbine

Savonius wind turbine

The Savonius wind turbine is a vertical-axis wind turbine (VAWT) that operates
using drag forces rather than lift. Developed by Finnish engineer Sigurd Savonius
in the 1920s, it features a simple design with two or more curved blades, often
forming an "S" shape when viewed from above. This configuration allows it to start
rotating easily, even at low wind speeds, providing high starting torque. Although
it is less efficient than lift-based turbines, the Savonius turbine is well-suited for
small-scale applications such as water pumping and ventilation. Its robustness,

40
low cost, and ability to perform in turbulent or variable wind conditions make it a
reliable choice for decentralized energy needs.

Figure 3.2: Savonius wind turbine D = 30 mm and D = 20 mm

Helicoidal wind turbine

The helicoidal wind turbine is a variation of the vertical-axis wind turbine
(VAWT), characterized by twisted or helical blades that form a smooth spiral shape.
This design helps reduce pulsating torque and vibrations, resulting in more stable
and quieter operation compared to traditional straight-blade turbines. Commonly
used in Savonius-type configurations, the helicoidal shape improves performance by
maintaining continuous interaction with the wind throughout the rotation. While
it operates primarily on drag forces and is less efficient than lift-based turbines,
the helicoidal wind turbine offers advantages such as simplicity, good self-starting
capabilities, and reliable operation in turbulent or low-speed wind conditions,
making it suitable for small-scale or urban installations.

Figure 3.3: Helicoidal wind turbine

41
3.4 Experimental Study of a Helical Savonius
3.4.1 Purpose of the Manipulation
The main goal of this experiment is to evaluate the aerodynamic performance
of a Savonius vertical-axis wind turbine by measuring the torque it generates and
the wind speed it is exposed to, under controlled conditions with a fixed fan speed
(constant wind source).

3.4.2 Experimental Procedure and Equipment

• Make sure the equipment is stable.
• Turn the knob to the ON position to switch on the display screen.
• Once the screen is on, click on “machine status”.
• Use the “motor setpoint” knob to set the fan speed between 500 rpm and
800 rpm.
• Open the electro-brake control from the display screen. Then set the button
to ON to start the machine and begin the experiment.
• The display screen instantly shows the torque value in Nm, the speed of wind
turbine rotation in RPM, and the wind speed supplied by the fan.
• For more accurate measurements, the device can detect instantaneous fluctu-
ations in torque, wind turbine speed, power coefficient, mechanical power,
and wind power.
• It is necessary to define certain parameters such as air density (ρ = 1.3 kg/m3 )
and cross-section A = D · H (in our case, A = 0.0582 m2 ).
• The results are recorded instantly using a USB drive in the form of an Excel
workbook.
• During the experiment, it is tempting to increase the torque value by manually
adjusting the knob.
• Once the maximum torque value is reached, the vertical-axis wind turbine
stops rotating.
• The fan is switched off by clicking the OFF button on the screen, ending the
experiment.
• The USB drive is removed and the results analyzed on Mr. Marwa’s computer.

42
Measuring Instruments
The Entire Experimental Set-Up

Figure 3.4: Equipment

The Wind Tunnel

The experimental set-up consists of an ASZ series centrifugal fan that draws in
ambient air and discharges it into a horizontal duct 31.5 cm in diameter and 145 cm
in length to avoid the development of a boundary layer on the wall. A honeycomb
grid is placed at the duct inlet to even out the flow and reduce turbulence levels.
The model should be placed at the pipe outlet without exceeding its dimensions.

Figure 3.5: Wind Tunnel

43
Electromagnetic Brake
The electromagnetic brake, which operates in the event of a voltage failure, is
supplied with alternating current in the AC series. It is designed so that the shaft
on which it is mounted is locked in the rest position by spring pressure, but is
released when the coil is energized.

Figure 3.6: Electromagnetic brake (Deserti Meccanica 2025)

3.4.3 Experimental Results

We obtain the following database in an Excel sheet which contains about 9200
measured values for Vfan = 500–550–600–650–700–725–750–800 m/s.
We use the following data:

• R = 100 mm = 0.1 m

• ρ = 1.3 kg/m3

• A = D · H = 0.0582 m2

• D = 300 mm = 0.3 m

• H = 194 mm = 0.194 m

3.4.4 Used Formulas (Mariem Lajnef, 2020)

Determination of Torque Coefficient

Td
CT = 1 2
(3.4.1)
2
ρRAU∞

44
Determining Rotation Speed ω

2πN
ω= (3.4.2)
60

Determination of Wind Power Pw

1
Pw = ρAU∞
3
(3.4.3)
2

Determination of Rotor Power Pr

Pr = Td · Ω (3.4.4)

Determination of Power Coefficient Cp

Pr
Cp = (3.4.5)
Pw

Determination of λ

Cp = CT · λ (3.4.6)

3.4.5 Curves
Using the table below, we draw the curves:
Curve of Cp :

Figure 3.7: Curve of Cp = f (λ)

45
 Cp λ V Ct
V = 4.7 0.07477 0.04768 200 0.00356
0.19007 0.06385 150 0.01213
0.15763 0.07867 120 0.01240
V = 6.75 0.15328 0.04617 200 0.00708
0.19812 0.06216 150 0.00651
0.08605 0.08226 120 0.00708
V = 7.6 0.13978 0.04784 200 0.00669
0.19793 0.06412 150 0.01291
0.17337 0.07771 120 0.01304

Table 3.4: Table representing different values of Cp, λ, V, and Ct for given
conditions.

It can be seen that the value of Cp increases for all the speeds tested. Super-
imposing the three curves on the same graph shows that the curves increase until
they reach a maximum value, and then decrease. It shows that for a minimum
speed equal to v = 7.6 m/s, the Cp = f (λ) curve offers maximum performance.
Curve of Ct:

Figure 3.8: Curve of Ct = f (λ)

The shape of this curve is not logical and it’s due to the experimental results
and the materials and conditions of the experiment.

46
3.4.6 Interpretation of the Results
The maximum efficiency of this Savonius for a specific speed V=4.7 m/s is
0.14082%.
The maximum efficiency of this Savonius for a specific speed V=6.75 m/s is
0.11469%.
The maximum efficiency of this Savonius for a specific speed V=7.6 m/s is 0.10859%.

3.5 Conclusion
This chapter demonstrates the different performences of the wind turbines with
lab conditions and their optimum values.

47
Chapter 4

Numerical Simulation

4.1 Introduction
In this section, we further investigate vertical-axis wind turbines using numerical
simulation. We perform several simulations to evaluate the turbine’s performance
in terms of torque and power coefficient. This chapter presents the simulation steps
and parametrs, and the results in sequential order.

4.1.1 Numerical Study of the Helical Vertical-Axis Wind

Turbine
4.1.2 Geometry of the Savonius
SolidWorks Simulation of the VAWT with the Required Dimensions:
I created a 3D model in SolidWorks to import it into ANSYS. Using STEP files
from SolidWorks, the geometry is transferred to ANSYS DesignModeler. This tool
is the starting point for 3D calculations.

48
 Figure 4.1: 3D simulation of the Savonius wind turbine in SolidWorks

We introduce geometry with the following parameters:

Table 4.1: Geometric parameters of the helical Savonius

Parameter Value
D 265 mm
H 312 mm
De 145 mm
A 0.176202 m2
N 2
Blade thickness 3 mm

Calculation domain The first step when creating a numerical simulation is

to define the computational domain. To account for rotor rotation, the calculation
domain is divided into two distinct fluid zones: an external fluid domain representing
the wind flow (represented by a box), and a rotating domain representing the turbine
rotor (represented by a cylinder). The rotor fluid area contains the rotating blades
of the wind turbine (excluding the plates) and the rotating shaft. The interfaces

49
between the rotor and external domains are defined as interface boundaries designed
for a moving mesh application.

Figure 4.2: The two domains used in the simulation

4.1.3 The Mesh

The calculation domain is imported into the ANSYS Mesh interface. ANSYS
Workbench automatically generates a mesh, whose refinement can be specified. For
our case, we used: A structured mesh with Max Face Size = 0.01 m.

Figure 4.3: Domain mesh generated with 0.01 m element size

50
 Domain Nodes Elements
Default Domain
537,499 508,860
Modified
Rotor 27,488 133,888
All Domains 564,987 642,748

Table 4.2: Domain nodes and elements

4.1.4 General Flow Conditions (Set-Up)

In this work, due to the complexity of the geometry and difficulty in maintaining
mesh quality in certain regions, we limited our study to steady-state simulations
with a rotating relative frame of reference. The air properties used in the software
for all simulations are as follows: For each simulation, a uniform velocity is applied
at the inlet, and an angular velocity is assigned.

Figure 4.4: Inlet and outlet conditions

Air density = 1.3 kg/m3

4.1.5 Calculation Convergence

We set the number of iterations to 100. Note that, in steady-state, the angular
velocity converges towards a more or less stable solution.

51
 Air veloc-
Rotor velocity
ity
4.7 m/s 120 rpm
150 rpm
200 rpm
6.75 m/s 120 rpm
150 rpm
200 rpm
7.6 m/s 120 rpm
150 rpm
200 rpm
Tested air and rotor velocities

Table 4.3: Air and rotor velocities for simulation

Figure 4.5: Calculation residuals for flow around the helical Savonius (V = 4.7
m/s)

52
 General simulation conditions
Software CFX
Inlet condition Velocity
Outlet condition Pressure
Turbulence model SST (k–ω)

Table 4.4: General simulation conditions

Simulations Performed on the Same Project

Figure 4.6: Different simulation conditions realized in the same project

4.1.6 Results and Discussion

Spatial Distribution of Post-Simulation Parameters
Wind Speed in the Study Area

53
 Figure 4.7: Velocity distribution for V = 4.7 m/s and λ = 0.07

Turbulence Around the Rotor

Figure 4.8: Turbulance distribution for V = 4.7 m/s and λ = 0.07

Blade Pressure

54
 Figure 4.9: Pressure around the blades for V = 4.7 m/s and λ = 0.07

From these figures, we can see that it is possible to observe the distribution
of flow speeds when using a Savonius turbine. With a uniform wind speed of 4.7
m/s , the air speed reaching the rotor is around 2.4 to 9.6 m/s. This means that
this type of turbine has a significant effect on the wind that attacks it upstream.
What’s more, the average air speed leaving the rotor upstream is around 3 to 5
m/s. Consequently, this turbine also has a significant impact on the downstream
wind. With regard to pressure, we can see that there are pressure values around
the rotor towards the walls of around 60Pa, and low values downstream of the
rotor. The rotor of a turbine can have a significant influence on wind pressure.
When air comes into contact with the rotor blades, it is forced to change direction
and pass through the space between the blades. This can lead to a drop in air
pressure downstream of the rotor, as some of the air’s kinetic energy is converted
into mechanical energy by the rotor. For a wind speed =4.7 m/s and for different
values of the angular velocity of the wind turbine, we obtain the curve of Cp below:

Figure 4.10: Curve of Cp

55
 It can be seen that the value of Cp increases for all the speeds tested.

Figure 4.11: Curve of Ct

Analytical calculation of angular velocity (omega) The value of (omega )

is an input for the Fluent software (Input). The expression for calculating angular
velocity is :
V ×λ
ω= (4.1.1)
R
R=0.1325m

4.2 Comparison Between Numerical and Experi-

mental Results
To confirm the validity of the model, it is necessary to compare the experimental
results obtained in the wind tunnel with the numerical results obtained using the
Ansys CFX calculation code.

Figure 4.12: Curve of Cpexp and Cpnum

Superimposing the experimental and numerical curves of Cp shows that there

is agreement between the values.

56
 It can be seen that the CP power coefficients obtained numerically are over-
estimated compared with those obtained experimentally. For the Ct curve , we
couldn’t compare them due to the experimental results due to the wrong results
obtained.
This discrepancy between numerical and experimental results is due to several
factors.Errors can occur at the meshing stage. Since the mesh is not very fine,
due to a lack of computer resources, the agreement between the results is not
very precise and therefore certain forces acting on the wind turbine are not well
quantified. This leads to an overestimation of power coefficient values.In addition,
in the experimental part, the errors derive from the fact that wind turbine speed
and torque measurements are recorded instantaneously. As a result, we’re faced
with averaging the measurements in order to show the results, and this leads
to uncertainty. What’s more, the 360-degree knob that varies the torque is not
graduated, so the torque is changed arbitrarily, leading to error.
Although the Cp against Lambda curves for the numerical research using Ansys
and the wind tunnel tests are similar and demonstrate a similar wind turbine
performance, this cannot ignore the differences in the curves values and slopes.

4.3 Conclusion
The numerical simulations of a steady-state flow for a vertical-axis helical
Savonius wind turbine and the results both have been presented in this chapter
and compared to the experimental results .
The primary goal of the numerical modeling was to examine the aerodynamic
performance of both conventional and helical Savonius wind turbines. We were able
to assess the wind turbines’ performance and approximate the flows around them
thanks to the numerical model we used. As a conclusion , this chapter includes a
verification of the numerical results via the numerical simulation and an explanation
of the error .

57
Conclusion

This research uses a computational fluid dynamics (CFD) approach to investigate

and assess the performance of vertical-axis Savonius wind turbines in Tunisia. The
study explores the benefits of this type of wind turbine and the CFD technique
used to simulate flows around them. A numerical model for both conventional
and helical Savonius turbines was created using SolidWorks software and ANSYS
Workbench. Numerical simulations were performed on the steady-state models to
describe the turbines’ operations. The models were validated by comparing their
results with experimental findings and evaluating the wind turbine’s performance.

“Software is a Great combination between artistry and engineering .” Bill Gates

.

58
Bibliographic References

[1] Statitica 2025.

[2] International Energy Agency. World Energy Outlook. 2022.
[3] Nada Kh. M. A. Alrikabi. Renewable Energy Types. Journal of Clean Energy
Technologies, Vol. 2, January 2014.
[4] Design Considerations of Rotor Configuration for Straight-Bladed Vertical
Axis Wind Turbines.
[5] Finite Volume Methods ,Robert Eymard , Thierry Gallou¨and Raphaele
Herbin January 2019.
[6] Fossil-Fuel-Consumption-Trend-and-Global-Warming-Scenario-Energy-Overview.
[7] Tunisia Country Commercial Guide. International Trade Administration.
https://www.trade.gov/country-commercial-guides/tunisia-electrical-power-systems-
and-renewable-energy, 2024.
[8] https://apnews.com/article/climate-change-renewable-energy-solar-wind-electricity-
6d570ec401b6762453ec3af0ce973694. In: ().
[9] https://energycapitalpower.com/tunisias-strategic-push-toward-renewable-energy-
growth/. In: (2025).
[10] https://www.eurus-energy.com/en/release/press-release/104283/. “eureus en-
ergy,Participation in 100MW Solar Photovoltaic IPP Projects in Tunisia”.
In: (August,2024).
[11] https://www.expadd.org/en/tunisia-in-sfax-a-factory-will-convert-household-
waste-into-methane/. In: (2024).
[12] https://www.miga.org/press-release/miga-boosts-tunisias-first-large-scale-solar-
energy-project. “MIGA Boosts Tunisia’s First Large-Scale Solar Energy
Project”. In: (July 31, 2024).
[13] Introduction à la CFD (Computational Fluid Dynamics) Bouhelal Abdelhamid,
Arezki Smaili , ENP 2021-2022.

59
[14] IRENA (2024), Green hydrogen strategy: A guide to design, International
Renewable Energy Agency, Abu Dhabi.
[15] Gye-Chun Cho J. Carlos Santamarina. “Energy Geotechnology”. In: (2013).
[16] École Polytechnique de Montréal] Jamati F. (2011). Étude numérique d’une
éolienne hybride asynchrone [Mémoire de maîtrise. MA thesis.
[17] SAMEER SAADOON AL-JUBOORI. Green Energy - An Introduction. Jan-
uary 2015.
[18] Mohamed R. Gomaa Mohammad A. Al-Rawajfeh. Comparison between hori-
zontal and vertical axis wind turbine. Tech. rep. Department of Mechanical
Engineering, Faculty of Engineering, Al-Hussein Bin Talal University, Ma’an,
Jordan, March 2023.
[19] Guide to Purchasing Green Power. Renewable Electricity, Renewable En-
ergy Certificates, and On-Site Renewable Generation. U.S. Environmental
Protection Agency, March 2010 , Updated: September 2018.
[20] Renewables 2024, Analysis and forecast to 2030. International Energy Agency.
[21] Reynolds-Averaged Navier–Stokes Equations for Turbulence Ansys CFX In-
troduction,Modeling,Giancarlo Alfonsi.
[22] Statistical Review of World Energy. 2024.
[23] CLEAN ENERGY PROJECT ANALYSIS: RETSCREEN ENGINEERING
CASES TEXTBOOK. WIND ENERGY PROJECT ANALYSIS CHAPTER.
Minister of Natural Resources Canada 2001 - 2005., 2005.
[24] ABDOUL KARIM TRAORÉ. “GESTION D’UN SYSTÈME AUTONOME
HYBRIDE PHOTOVOLTAÏQUE ÉOLIEN POUR APPLICATIONS AGRI-
COLES”. PhD thesis. L’UN1VERSITÉ DU QUÉBEC À TROIS-RIVIÈRES,
JUILLET 2016.
[25] “Yale departement of astronomy”. In: ().

60