Jeannine Uwibambe PDF

Design of Photovoltaic System for Rural
Electrification in Rwanda
by
Jeannine Uwibambe
Supervisor:
Professor Hans Georg Beyer
University of Agder, 2017

Faculty of Engineering and Science
Department of Renewable Energy
University of Agder, Norway
Design of Photovoltaic System for Rural Electrification in Rwanda
Abstract
In this century of accelerated development in various domains, some African countries are still
facing a challenge of lack of power due to its scarce in some places, where by the main source
of power (generation of electricity) is hydro since thermal and fuel are still on a small scale.
This problem results in less productivity and economic decline of some countries like Rwanda
which is among African countries that are at a very high speed in development, the grid lines
from distant places are stack and they are few compared to the need of electricity in all corners
of the country, especially in rural areas whereby each household needs power usage instead of
using local and traditional means of ironing and lighting at home. This issue can be solved
using other alternative sources of Renewable Energy for rural electrification such as
Photovoltaic systems. Therefore, this master’s thesis project is mainly focusing on the design
of off-grid Photovoltaic systems that include an economic evaluation between the use of an
individual solar home system of 200W and a village PV system of 10kW so that the satisfactory
of people and the targets of the country can be easily achieved. Under this Master’s thesis work,
the first part is focused on the analysis of electricity consumption based on single house owning
individual solar home systems taking a case study of one village in Rwanda called Kanazi
located in Eastern Province, Bugesera District, Nyamata Sector. This analysis is done through
the evaluation of the average primary load profile in consecutive number of hours per day
depending on the PV production capacity.
The purpose of this analysis is to obtain the optimum sizing of the PV panel as well as the
battery capacity that can be used for providing electricity to households. The second step is to
design a village PV system with a big battery and inverter that can generate electricity for the
selected village depending on the estimated the average daily load profile for a typical single
house in Kanazi village. Finally, both proposed and designed systems will be compared to
obtain a system that is more reliable and economical for electricity production. In this project
the design and simulation tasks will be performed through the help of Homer software and
Excel Software. The information about the average monthly solar radiation on the selected site
and the characteristics of PV system components will be provided by different internet
websites, PVGIS (Photovoltaic Geographical Information System) for Africa, different books,
scientific research papers, journals and the field survey that have been conducted.
Keywords: Rural electrification, Renewable Energy, Off-grid PV Systems, Grid lines, solar
home systems, mini-grid systems and Homer software.
i
Dedication
To my beloved Son Prince Daniel IGANZE
ii
Acknowledgements
First and foremost, I would like to give thanks to the Almighty God who has walked with me
throughout this journey and before. Without the constant guidance and protection of the Lord,
this work would barely be a dream.
Special thanks to my supervisor Professor Hans-Georg Beyer. Professor Beyer has been more
than an academic supervisor. This Professor kindly pushed me to do the best work, patiently
guiding me throughout each stage. This thesis would not have been possible without the
tenacity with which he oversaw this work. I am especially thankful that Professor Beyer was
understanding during the rough patches of this journey.
Finally, I am greatly grateful to my husband, my daughter, my son, my parents, my siblings
who have always encouraged me and provided me with moral and spiritual support. I received
great inspiration from their love and companionship, even from thousands of miles away.
iii
Table of Contents
Abstract ....................................................................................................................................... i
Dedication ..................................................................................................................................ii
Acknowledgements .................................................................................................................. iii
List of Figures ..........................................................................................................................vii
List of Tables ............................................................................................................................ ix
List of Abbreviations ................................................................................................................. x
1 Introduction ........................................................................................................................ 1
1.1 Background and Motivation ........................................................................................ 1
1.2 Problem Statement ...................................................................................................... 1
1.3 Goals and Objectives ................................................................................................... 2
1.4 Research Methods ....................................................................................................... 3
1.5 Key Assumptions and Limitations .............................................................................. 3
1.6 Outline ......................................................................................................................... 4
2 Literature Review............................................................................................................... 5
2.1 Integrated Renewable Energies in Rwanda ................................................................. 5
2.1.1 Hydro Power generation ...................................................................................... 5
2.1.2 Geothermal energy resources ............................................................................... 6
2.1.3 Wind power .......................................................................................................... 7
2.1.4 Methane Gas ........................................................................................................ 7
2.1.5 Peat to power........................................................................................................ 8
2.1.6 Biogas power ....................................................................................................... 8
2.1.7 Solar energy ......................................................................................................... 8
2.2 Basic operation mechanism of the photovoltaic cell ................................................... 9
2.3 Solar Home Systems ................................................................................................. 10
2.3.1 The Barefoot power pack of 5W micro-kit ........................................................ 12
2.3.2 BBOXX17 of 50W Solar home system ............................................................. 12
2.4 Mini-grid Systems ..................................................................................................... 13
2.5 Overview on HOMER Software ............................................................................... 17
3 Data Collection and selected site ..................................................................................... 19
3.1 Introduction ............................................................................................................... 19
3.2 Rwanda Energy Sector Overview ............................................................................. 20
3.3 Data for Bugesera District ......................................................................................... 20
iv
3.4 Single house load profile ........................................................................................... 24

3.5 Solar resources analysis for Nyamata Sector ............................................................ 26
3.6 Additional resources for the selected site .................................................................. 28
4 PV System Components behaviors and Costs ................................................................. 29
4.1 Solar Panels ............................................................................................................... 30
4.1.1 Electrical characteristics of a solar panel ........................................................... 30
4.1.2 Effect of Temperature on the performance of a Solar panel .............................. 35
4.1.3 Effect of irradiance on the performance of a Solar panel .................................. 36
4.2 Battery storage........................................................................................................... 37
4.3 Solar Inverter and MPPT System .............................................................................. 40
4.4 Other Devices ............................................................................................................ 41
4.5 Cost Estimation for Solar Panels, Battery storage and Inverter ................................ 41
5 PV Systems Modelling .................................................................................................... 43
5.1 Design of 0.2kW Solar Home System ...................................................................... 43
5.1.1 Primary Load Input ............................................................................................ 44
5.1.2 PV inputs............................................................................................................ 45
5.1.3 Battery inputs ..................................................................................................... 46
5.1.4 Modelling of Solar resource inputs .................................................................... 47
5.2 Design of 10kW off-grid PV System for the village ................................................. 50
5.2.1 Battery input....................................................................................................... 51
5.2.2 Converter input .................................................................................................. 53
5.2.3 Economic inputs................................................................................................. 54
6 Results .............................................................................................................................. 56
6.1 Results for Solar Home System ................................................................................ 56
6.1.1 Overall Optimization Results............................................................................. 57
6.1.2 Sensitivity Results .............................................................................................. 58
6.2 Results for 10kW Village PV System ....................................................................... 63
7 Discussion ........................................................................................................................ 69
8 Conclusion and Future work ............................................................................................ 71
9 References ........................................................................................................................ 72
Appendices ............................................................................................................................... 76
Appendix A Mobisol customers in Kanazi cell, Nyamata sector, Bugesera district. ........ 76
Appendix B Photovoltaic Geographical Information for Nyamata Sector. ....................... 77
v
Appendix C Details on daily consumption for Kanazi village. ......................................... 79

Appendix D Hourly power distribution per day for selected houses. ................................ 83
vi
List of Figures
Fig. 2.1: Global horizontal irradiation map for different districts of Rwanda [17]. .................. 9
Fig. 2.2: Basic operation mechanism of the solar photovoltaic cell [12]. ............................... 10
Fig. 2.3: A typical Solar home system installation on the roof of a house in Nyamata. [13] .. 11
Fig. 2.4: Single line diagram of the basic Solar Home System in Kanazi village. .................. 11
Fig. 2.5: The Barefoot power pack of 5W micro-kit used for SHS applications. [5] .............. 12
Fig. 2.6: BBOXX17 of 50W Solar home system used for rural electrification purposes. [5] . 12
Fig. 2.7: Main Energy Sources in Rwanda [15]....................................................................... 13
Fig. 2.8: utility-scale of 8.5MW PV power plant constructed in Agahozo-Shalom Youth Village
in Rwanda [9]........................................................................................................................... 15
Fig. 2.9: Rwanda's total production in terms of energy generated by various sources [16]. ... 16
Fig. 2.10: Rwanda's total energy consumption through public and private services [16]. ...... 17
Fig. 3.1: Rwanda administrative map with a clear illustration of all provinces and their districts.
[22] ........................................................................................................................................... 19
Fig. 3.2: Rwanda electricity profile showing the current installed capacity and rural access to
electricity [24]. ......................................................................................................................... 20
Fig. 3.3: Monthly average values of various climate data for Bugesera District. [26] ............ 21
Fig. 3.4: Administrative Map of Bugesera District with all different sectors and rivers. [25] 22
Fig. 3.5: Map of Geography allocation of Nyamata Sector in Bugesera district. [28] ............ 23
Fig. 3.6: View of Kanazi village of Nyamata Sector showing the topography. [29]............... 23
Fig. 3.7: A Typical house for the poor family in Kanazi cell of Bugesera district. [31] ......... 24
Fig. 3.8: Average daily load profile for a typical single house in Kanazi village selected for
case study. ................................................................................................................................ 26
Fig. 3.9: Monthly global radiation for the selected village in Nyamata, from Homer Software.
.................................................................................................................................................. 26
Fig. 4.1: General diagram of the PV system showing the main components. [3] ................... 29
Fig. 4.2:The I-V and Power Curves for the Solar Panel [38]. ................................................. 31
Fig. 4.3: I-V Curve and ratings of Solar Panel designed to be connected to a 12V battery [39].
.................................................................................................................................................. 32
Fig. 4.4: Simplified Equivalent Circuit Model for a Photovoltaic Cell [38]. .......................... 33
Fig. 4.5: Effect of series (a) and shunt(b) resistance on I-V characteristics of the PV Cell [41].
.................................................................................................................................................. 34
Fig. 4.6: Getting the Fill Factor from the I-V Curve [38]. ....................................................... 34
Fig. 4.7: Temperature Effect on I-V Curve [38]. ..................................................................... 35
Fig. 4.8: The effect of Irradiance on I-V curve of a solar panel [46]....................................... 36
Fig. 4.9: Vision 6FM55D, 12V, 55Ah Battery storage [49]. ................................................... 38
Fig. 4.10: Surrette battery 6CS25P, 6V. [50] ........................................................................... 38
Fig. 4.11: Capacity curve of the Surrette battery 6CS25P, 6V, from Homer. ......................... 39
Fig. 4.12: Life cycle curve for Surrette battery 6CS25P, 6V, from Homer. ............................ 39
Fig. 4.13: Inverter's circuit diagram with different electronic switching circuits [34]. ........... 40
Fig. 5.1: Single line diagram of the basic Solar Home System in Kanazi village. .................. 43
Fig. 5.2: Inputs for modelling a solar home system of a single house in HOMER. ................ 44
Fig. 5.3: Input window for primary load of a typical single house in Homer. ........................ 45
vii
Fig. 5.4: Homer Input window for PV system used for electricity generation for a house in
Kanazi cell. .............................................................................................................................. 46
Fig. 5.5: Input window for the battery selected for solar home system design in Homer. ...... 47
Fig. 5.6: Resource window with data for Kanazi village taken from PVGIS entered [19]. .... 48
Fig. 5.7: Yearly scaled data for solar radiation intensity in Kanazi village based on hourly load.
.................................................................................................................................................. 49
Fig. 5.8: Single line diagram for a village PV system of 10kW capacity in kanazi cell. ........ 50
Fig. 5.9: Inputs for modelling a 10kW PV system for kanazi village using Homer Software.51
Fig. 5.10: Battery storage window with properties in homer component library. ................... 52
Fig. 5.11:Battery input window, from homer showing properties and cost curve................... 53
Fig. 5.12: Converter input window, from homer showing the cost curve and different sizes for
the converter............................................................................................................................. 54
Fig. 6.1: Simulation results based on cash flow summary and Net Present Cost categorized by
component. ............................................................................................................................... 56
Fig. 6.2: Levelized cost of energy vs PV array capacity with a fixed renewable fraction of 10%.
.................................................................................................................................................. 61
Fig. 6.3: Graphical Sensitivity Results based on Levelized COE............................................ 62
Fig. 6.4: Simulation results based on cash flow summary and Net Present Cost categorized by
component ................................................................................................................................ 63
Fig. 6.5: Graphical Sensitivity Results based on LCOE: ......................................................... 66
Fig. 6.6: LCOE vs PV array capacity with a fixed renewable fraction of 10% for the village
system of 10kW. ...................................................................................................................... 68
viii
List of Tables
Table 2.1: Total number of solar power plants and total capacity [9]. .................................... 16
Table 2.2: Electricity tariff for residential and non-residential consumption in Rwanda [17].
.................................................................................................................................................. 17
Table 3.1: Daily load evaluation for a poor single house in Kanazi village. ........................... 25
Table 3.2: Monthly global radiation for the selected village in Nyamata sector, from Homer
tool. .......................................................................................................................................... 27
Table 3.3: Average monthly values for other climatic parameters in Nyamata Sector. [33] .. 28
Table 4.1: Output Parameters of Solar Cell Under Different Temperatures [40] .................... 36
Table 4.2: Range of Cost Estimation for monocrystalline Solar Panels depending on the system
size [44].................................................................................................................................... 42
Table 5.1: Economic inputs used by Homer for calculating the NPC for the system. ............ 55
Table 6.1: Overall Optimization Results without shortage. ..................................................... 57
Table 6.2: Overall Optimization Results with Shortage. ......................................................... 58
Table 6.3: Tabular Sensitivity Results without shortage. ........................................................ 59
Table 6.4: Tabular Sensitivity Results with shortage. ............................................................. 60
Table 6.5: Overall Optimization Results with no Shortage. .................................................... 64
Table 6.6: Categorized Optimization Results with zero shortage............................................ 64
Table 6.7: Overall Optimization Results with Capacity shortage............................................ 65
Table 6.8: Categorized Optimization Results with capacity shortage ..................................... 65
Table 6.9: Tabular Sensitivity Results with shortage. ............................................................. 67
ix
List of Abbreviations
EDPRS Economic Development and Poverty Reduction Strategy
EICV3 Third Integrated Household Living Conditions Survey
HOMER Hybrid Optimization Model for Electric Renewables
ESMAP Energy Sector Management Assistance Program
PVGIS Photovoltaic Geographical Information System
IRENA International Renewable Energy Agency
NREL National Renewable Energy Laboratory
NGOs Non-Governmental Organizations
MPPT Maximum Power Point Tracker
IED Innovation Energy Development
DDP District Development Plan
SHS Solar Home System
AC Alternating Current
NPV Net Present Value
NPC Net Present Cost
DC Direct Current
PV Photovoltaic
x
1 Introduction
The application of renewable energies contributes to global warming prevention and as a matter
of fact photovoltaic systems have been increasingly developed in recent years due to the global
benefit of natural resources conservation. It is also evident that fossil fuel-based energy sources
will be depleted over time since they are finite and consequently they have been proven to
contribute to global climate change. To protect our environment and increase electricity access
in remote areas, green and clean energy alternatives like solar energy, absorbed by photovoltaic
systems can be of great importance.
1.1 Background and Motivation

In Rwanda, there is a serious problem of electricity access especially in rural areas, this is very
crucial in affecting sustainable development of the country. The current situation shows that the
grid connected is estimated to be around 23%, whereby rural villages that are connected to the
national grid accounts for only 5% and in addition, statistics shows that 85% of Rwandan
population live in rural areas while only 15% accounts for urban citizens. [1] [2]
The most common activity observed in these areas is farming for food provision and other life
basic needs security. For the case of Rwanda with many population in rural areas, there is a
challenge of energy extension and development in other economic sectors.
The topology of the electric grid in Rwanda is another important aspect. There is presently
insufficient electrical power to compensate for electricity demand in Rwanda, most of the power
produced from different power plants is distributed to urban areas and business centers. The
power supply is done using single lines because the transmission network is very radial in
nature. Grid extension is affected by economic constraints such as high cost of electricity that
is not affordable for rural consumers as well as geographical conditions, and therefore, it’s hard
for poor people living in far distances from grid lines to get power.
In fact, there is a lack of alternate paths for electricity in transmission network and notably the
power service related to rural areas and this has a negative impact of pushing village residents
to move in cities. This vulnerability of rural areas is the major purpose of this thesis work for
providing the more reliable and typical solution for rural electrification in Rwanda.
1.2 Problem Statement

Electrical power access plays the vital role in accelerating economic development by improving
health and life standards. Significant investments have been made in energy sector to improve
electricity access in households and more efforts need to be made by Rwandan government in
collaboration with private sectors to meet the sets target of electrifying 70% of households by
2017/18 and 100% by 2020. [1]
For the above targets to be accomplished, a combination of various solutions that focus on the
geographical location, income and consumption level is needed instead of using the traditional
1
connection to the grid that may not be suitable for all rural households. Off grid PV systems can
be proposed as a sustainable solution to be implemented with a wide range of technologies from
a basic solar home system that can produce electricity for the single house with a radio, mobile
phone and four lamps to stand-alone systems that can generate high levels of electrical power
that can be used by both the households and other business centers.
This approach can serve as an economical and reliable solution for increasing the rate of
electricity access by the entire rural community as well as for the country in general [1]. In
recognition of the above proposed solutions, the research had also proven that the long term
annual average global irradiation in different districts is above 1700 kWh/m2. This shows how
many locations of Rwanda are favorable candidates for application of PV solar systems [2].
In this regards, the solar home system and stand-alone system in Eastern Province, Bugesera
District, specifically in Nyamata Sector, can bring expected positive benefits for households
and public services since many villages in this site people do not have access to national electric
grid line. This low accessibility is caused by high cost of transmission line per km which is
around $15000 for low voltage and $30400 for medium voltage [1] [2]. Therefore, since the
selected place is dry with high intensity of sun radiations, this makes the site more attractive for
the use of PV systems for electricity generation and storage.
1.3 Goals and Objectives

The main goal of this project work is to show how a photovoltaic system can be used to solve
the problem of electricity access in Kanazi, the village located in Nyamata sector of Bugesera
District. Since electricity from the grid is not easily available in this village due to the high cost
of transmission lines per km, photovoltaic technology such as solar home system and stand-
alone solar systems are proposed in terms of cost and efficiency to generate electricity for
households and public service applications. Due to climatic conditions, Nyamata sector receives
abundant amounts of sunshine all year around. Since solar energy is available only during the
day time, it is important to use it with energy storage device like battery to supply the load
during night hours to build a self-sufficient system. These approaches will compensate for
electricity demand hence contributing to sustainable and economic development of the country.
The following objectives should be accomplished to achieve the desired goal:
 Analysis of monthly solar irradiation within Nyamata sector in bugesera district of the
eastern province of Rwanda.
 Estimation of average daily load profile at the selected site.
 Proposing a basic solar home system layout and stand-alone system configuration.
 Component selection and cost analysis.
 Systems modelling and simulation by using HOMER software.
 Performance evaluation of optimal photovoltaic system.
 Cost comparison between the chosen system and the existing Solar home systems.
2
1.4 Research Methods

This project work is mainly summarized in several steps and procedures combined with the data
collection process based on renewable energy resources. These include the knowledge of village
load profile, a clear understanding of photovoltaic system configurations, study of component
behavior as well as cost, modelling and simulation of the solar home system as well as the stand-
alone solar system, selection of optimum system referring to the simulation results and
performance analysis between the proposed individual solar home system and the stand-alone
system for the village.
Basically, the preliminary step was the determination of daily load profile for the selected
village. Due to the equatorial location of Rwanda, there are no specified season for winter or
summer. Therefore, the case of load profile variations caused by change of season is not
considered. In fact, the knowledge of the average daily load profile for a typical single house in
Kanazi village was obtained through the self-performed field survey during summer holidays,
since my family is in the selected site it was easy for me to get important data from many
villages.
Furthermore, the data obtained from grid connected rural villages will enable me to do the
design for the selected site that do not have access to electricity from the grid. In addition,
various parameters such as, family classes based on the number of equipment, the number of
households and public services will be considered. Nevertheless, a reasonable assumption can
be useful for estimation of the load curve in case of unavailability of specific data for the
selected site.
Finally, Homer software will be used for design, modeling and simulation of the entire system
to get and analyze the possible profitable solutions.
1.5 Key Assumptions and Limitations

The scope of this project work is limited to the determination of optimal photovoltaic system to
generate electricity for rural areas. In addition, a performance evaluation of the system has been
done by taking into consideration the following assumptions.
 The important information on the irradiance at the selected site were derived by
PVGIS for Africa and the data from this site are helpful for providing the solar
resources input to be used for modelling and simulation of solar PV systems
 The Annual solar resources input and the primary load profiles are assumed to
remain constant throughout the project lifetime.
 This project life time was estimated based on the guarantee of PV panels which is
mentioned to be around 25 years.
 The daily load cycles for consumers were also assumed to be the same because the
temperature variations at the selected site are approximately fixed because summer
and winter seasons are not distinct.
3
The final designed system has the following limitations.

 The issues of PV system stability and control will not be included in this research
project.
 Design of Rwanda electric grid will not be part of this project work.
 Only Homer software will be used for modelling and simulation of the entire system.
1.6 Outline
This project details the design and simulation of a photovoltaic system in Rwanda for increasing
electricity access in rural areas especially focusing in one selected village named Kanazi,
located in eastern province, bugesera district, Nyamata sector. Therefore, the work is organized
as follows.
Chapter 2 provides a brief literature review about Rwanda, focusing on solar resources, rural
electrification strategic plan in general and the application of photovoltaic systems for
electricity generation by going into details of different PV system types and their functionalities
as well as homer software description.
Chapter 3 shows different data collected in the selected site with a clear presentation of the
average daily load profile for a typical single house in Kanazi village selected for case study
and available natural resources within the boundaries of the chosen district.
Having the knowledge of both the proposed system sizes to be modelled and site location of the
selected village, Chapter 4 will be concerned with the overview of the major components used
by a basic solar home system and isolated mini-grid village system technologies. It illustrates
the important characteristics of the system components such as electrical characteristics,
operation and cost implications.
Chapter 5 describes different designs of a basic solar home system as well as a stand-alone
village system. Various components that compose both mentioned systems will be modelled.
In this chapter, the behavior and cost estimation of different components that compose a typical
PV system are explored and Homer software is used for modelling the entire system.
Chapter 6 illustrates the results obtained from the simulation of both types of PV systems and
Chapter 7 discusses about different results obtained for achieving the least cost feasible options.
Chapter 8 summarizes the conclusions from this research project, and gives suggestions for
better future work. Finally, a list of bibliography used, appendices of excel sheets and field
survey as well as PVGIS data are shown at the end of the project report.
4
2 Literature Review
Renewable energy which include solar energy insures a clean energy supply and it has been
considered to play an important role of global warming prevention by fighting against climate
change [3]. Solar Energy is mainly generated depending on the intensity of the sun rays reaching
the solar panel and their wavelengths. [4] Electrification by different solar technologies had
been deployed in Rwanda since 1980’s mainly through support from donors and Non-
Governmental Organizations (NGOs) and a study conducted by ESMAP (Energy Sector
Management Assistance Program)/World Bank project in 1991 provided an assessment of the
market at that time. [5]
The use of solar energy for electricity generation is a non-consumptive use of a natural resource
and consumes no fuel for continuing operation since renewable energy is a clean source of
energy with the potential to contribute greatly to a more ecologically, socially, and economically
sustainable future. Nevertheless, the major challenge in solar power plant is to maximize the
wavelength of sun rays and minimize the effect of temperature on solar cells. [4] Solar energy
can be made more economical by reducing investment and operating costs and by increasing
solar plant performance and the most significant cost reductions are likely to come from
innovations in solar field design, which could bring down the Levelized cost of energy (LCOE)
by 15% to 28%, depending on the technology. [6]
Furthermore, renewables in combination with batteries allow stand-alone operations and
batteries are now a standard component of solar PV lighting systems and solar home systems.
The impact of off-grid renewable energy systems will not only be measured in terms of their
usage or reduced costs for electricity consumption in rural areas, but also in the context of their
effect on the lives of some 1.16 billion people who today are totally without access to electricity.
[7]
Regardless of ambitious electrification programs established by the Rwandan government,
more than 1 million households will likely still require off-grid electricity by 2020 and Survey
results show that average off-grid household energy expenditures in Rwanda are around RWF
500 or US$0.80 per week. [5]
2.1 Integrated Renewable Energies in Rwanda

Rwanda has a range of indigenous resources that complement each other in the energy mix, and
the government has a plan of utilizing these to reach its target of increasing total capacity by
80% in the year 2017.This will help to diversify away from fossil fuels due to the related high
cost of generation [8]. The following section outlines various physical potential of existing
renewable energy resources.
2.1.1 Hydro Power generation

Different Studies prove that Rwanda has significant potential for hydropower generation since
Rwanda’s topography is most suitable for medium to high head pico and micro hydro run-of-
river schemes [9]. The country possesses many rivers and streams which are not yet exploited
5
but there are 333 potential sites discovered across many locations and additional viable sites
have already been, and are likely to continue to be identified.
While current installed capacity stands at 59 MW, the government estimates untapped overall
technical hydropower potential of over 300 MW which varies according to different studies
made at different times [9] [8]. The Rwanda Hydropower Atlas conducted by the Rwanda
Ministry of Infrastructure in 2009, found that many potentially feasible sites would be rated
between 50 kW and 1 MW in potential capacity. [8] In Rwanda, there are 333 possible sites
identified for micro hydropower generation and these sites are located along the major rivers
flowing south from Mount Karisimbi in the north of the country, and along the Ruzizi River
towards Lake Kivu in the west. Larger projects are already under development in cooperation
with Rwanda’s neighbors.
The study estimated a potential of 96 MW for the category of micro-hydro projects.12 An
assessment of the energy sector undertaken by the African Development Bank in 2013 estimated
Rwanda’s domestic hydropower potential at 313 MW, broken down into 130 MW of domestic
and 183 MW of regional hydro resources. Feasibility studies have been completed or are under
way for many of the sites, representing at least 32 MW of technically viable new capacity. [9]
In addition, over 192 potential Pico hydro sites (with potential capacity of less than 50 kW)
have been identified and Feasibility studies and assessments are being carried out on an ongoing
basis, with the result that the amount of technically viable new capacity seems to increase
constantly. [8]
Evidently, it would be valuable to have a more detailed resource mapping for the hydropower
sector, especially taking a spatial river basin approach before giving priority to specific sites for
development. This approach was adopted by an on-going comprehensive assessment of hydro
resources on the Akanyaru River basin, located on the border between Rwanda and Burundi.
These resources can be developed in cascade form, with 11 domestic sites and 3 shared sites
recommended for further feasibility analysis. Based on preliminary assumptions, these projects
could potentially increase Rwanda’s total installed capacity by over 25 MW [9].
2.1.2 Geothermal energy resources

Rwanda may have geothermal potential which is manifested in the form of hot springs.
Potential geothermal resources have been categorized into four main prospect areas, all within
the belt along Lake Kivu: Karisimbi, Kinigi, Gisenyi and Bugarama. The potential of
geothermal energy is still uncertain. Exploration studies have estimated the potential for
commercial power generation to be in the range of 170 to 340 MW [8]. Given the complexity
involved in determining the commercial viability of geothermal power, however, much more
detailed exploration studies and sub-surface drilling are required [9]. Two preliminary drilling
projects have not provided proof of this potential, but as geothermal exploration is a relatively
long-term process, more studies are needed to assess and confirm potential.
6
2.1.3 Wind power

Being located close to the equator, Rwanda’s inherent resource potential for wind energy is low.
Wind power potential was evaluated in a rapid wind energy resource assessment carried out in
five locations in Rwanda over the course of 2011. [8] As results, the preliminary indications
from the analysis of recorded field measurements of wind speeds and climate data were that
most of Rwanda is not highly suitable for wind energy. The Eastern province was identified as
the location with the most promising potential, and a simple analysis comparing wind and solar
energy feasibility suggested that wind energy could be competitive in this region. [9] However,
more feasibility studies and assessments are needed for this to be precisely determined. Another
academic study using modelling analyses based on recorded wind measurements and power
density at selected Rwandan meteorological stations noted that electricity production around
Gisenyi station could be possible with a good mean value of both wind speed and power density,
whereas in areas such as Kigali, Butare and Kamembe, wind energy potential is only sufficient
for windmills or water pumping for agricultural and intuitional needs [8]. More detailed
resource assessments and feasibility studies are required to determine commercially viable wind
energy potential in Rwanda.
2.1.4 Methane Gas

Much of the country’s methane gas resources stem from the globally unique geology of Lake
Kivu and the naturally regenerating methane gas that is found there. Some 1000 studies have
been carried out, the consensus of which is that Lake Kivu is estimated to contain around 55
billion cubic metres of methane gas, with a further 150 to 250 million cubic metres of methane
being generated annually [9] .In recent years, Government has put great efforts on proving the
technical and commercial viability of safely extracting methane from Lake Kivu to produce
power. Indeed, the more methane gas that is extracted, the safer are riparian communities, as
there will be less methane gas that could potentially explode in case of a sub-surface volcanic
eruption. According to feasibility studies undertaken [by EWSA], it is possible to sustainably
extract sufficient gas to generate around 700MW of power [9]. This would put the national
potential at 350MW, as the resources are shared equally by Rwanda and DRC through an
international agreement.
A small pilot plant, owned and operated by Kibuye Power (KP I), and having an installed
capacity of 3.6MW, has been operational for some time [9]. The expected commissioning of
the 25MW KivuWatt power plant in late 2014 will go a long way, however, toward
corroborating the commercial feasibility of Lake Kivu methane gas as resource for large-scale
power generation. In addition to Lake Kivu, Rwanda also possesses inherent methane resources
that can be extracted from underground mining seams. Methane gas has a variety of commercial
and industrial uses, beyond the production of electricity, although it is currently anticipated that
the primary end-use of the resource will be for electricity. Feasibility studies for direct use of
methane gas in heating applications, fertilizer, and petrochemical production have already been
conducted and more are to be expected with a lower risk of extraction of the resource [9].
7
2.1.5 Peat to power

Currently peat is used as an input in cottage industries, cement production, and as a cooking
fuel in a small number of decentralized institutions. Rwanda has estimated reserves of 155
million tons of dry peat14 spread over about 50,000 hectares in Akanyaru, Nyabarongo,
Rwabusoro and other areas [9]. Approximately one-third of this resource is currently
commercially extractable for industrial heat or electric power production. These estimates are
largely based upon a ‘high level’ master plan undertaken in 1993, with only a small number of
samples having been taken and most work being carried out as desk research. In late 2013 and
early 2014, REG undertook more detailed resource assessments to complement and refine the
existing peat master plan.
Parallel to this, several projects have independently developed feasibility studies based on
resource potentials identified in specific locations. However, it is important that at the national
level Rwanda has an accurate assessment of its resources to determine how to use them most
efficiently and sustainably, as peat energy is not generally considered to be renewable. In
addition, some peat resources are located in environmentally sensitive habitats [9].
2.1.6 Biogas power

Biogas can be generated from a variety of biomass resources, including agricultural, human,
and municipal waste. The national resource potential for these alternative sources of methane
or biogas has never been systematically estimated through surveys or extensive resource
assessments. Pre-feasibility studies have been conducted for a landfill gas-to-electricity project
in Kigali, however, which suggested that commercial plant could be feasible in the near-term
future given higher population growth and volumes of waste collected and more systematic
separation of waste streams [9].
2.1.7 Solar energy

Rwanda’s solar radiation and solar resources were assessed by the U.S. National Air and Space
Agency (NASA) as well as the University of Rwanda. Rwanda’s Eastern Province has the
greatest potential for generating energy from solar resources. Another academic assessment,
undertaken in partnership with the MININFRA Department of Meteorology in 2007, used a
meteorological data set to estimate monthly averaged global solar radiation [9]. Rwanda’s
daily solar irradiation ranges from 4 kWh/m² north of the city of Ruhengeri to 5.4 kWh/m² south
of the capital, Kigali, in the Southern and Eastern provinces. However, conditions vary from
season to season, with average daily irradiation levels in the cloudy reaching about 4.5 kWh/m²
and the total annual potential is estimated to be around 66.8 TWh [9].
8
Fig. 2.1: Global horizontal irradiation map for different districts of Rwanda [17].
Rwanda is well benefited with solar energy, even during the months of the rainy seasons there
is daily and sufficient sunshine especially in the Eastern province which is known for high
irradiance values as it is indicated on the above Figure 2.1, the average daily global solar
irradiation on the tilted surface has been estimated to be 5.2 kWh per m2 per day from
Photovoltaic Geography Information System (PVGIS) [9]. The long-term monthly average
daily global irradiation range from 4.8 kWh/ (m2 day) (location Burera, month of May) of to
5.8 kWh/ (m2 day) (location Nyanza, month of July) which indicates a good potential for solar
energy development [9].
2.2 Basic operation mechanism of the photovoltaic cell

The use of PV Systems is the most effective and easy way to harness the solar energy within
the building environment [10]. PV facility uses semiconductors (PV cells) which absorb solar
energy to produce electricity through the photovoltaic effect. This physical process is a result
of the fact that some materials (i.e. silicon) produce electric current when exposed to light.
Sunlight is composed of photons or packets of energy and when these photons strike a PV cell
9
they may be reflected or absorbed or they may pass light through. When a photon is absorbed,
its energy is transferred to an electron in an atom of the semiconductor. Thereafter, the electron
can escape from its normal position associated with that atom to become part of the current in
an electrical circuit.
Special electrical properties of the solar cell provide the voltage needed to drive the current
through an external load. The electrical efficiency of the solar cell is obtained by firstly
determining the power generated by the cell at its maximum power point and the PV efficiency
(ƞ) is determined as the fraction between the maximum power which a solar cell can convert
from absorbed light to electricity (Pmax) and the incident power (Pin) as shown in the equations
below [10] [11].
𝑃𝑚𝑎𝑥 = 𝐼𝑚𝑝 ×𝑉𝑚𝑝 (2.1)
𝑃𝑚𝑎𝑥
𝜂= (2.2)
𝑃𝑖𝑛
Fig. 2.2: Basic operation mechanism of the solar photovoltaic cell [12].
2.3 Solar Home Systems

Solar Home Systems(SHS) are systems bigger than 15W that are sometimes sold as plug-and-
play systems, but more commonly as separate components, which need to be installed by a
technician [5]. Given the large size of these systems, proper estimation of the power demand,
sizing of the system components and a professional installation and maintenance is important.
That is, more consumer education is necessary to ensure proper use of the product. Also, after-
sales support may be necessary if one of the components (e.g. the battery) breaks down [5].
10
Fig. 2.3: A typical Solar home system installation on the roof of a house in Nyamata. [13]
The basic solar home system is represented by the single line diagram shown in Figure 2.4below
and this system is used for supplying electricity to residential consumers that use only DC loads
such as light bulbs, radio and mobile phones and in this case, they are supplied by a DC bus of
12V since we have considered one battery for the single house analysis in Kanazi village.
Fig. 2.4: Single line diagram of the basic Solar Home System in Kanazi village.
The Rwandan solar PV market is an early-stage market of small players that is poorly integrated
into the global and regional solar energy industry. But the Rwandan Government is working
with partners such as the European Union, the World Bank and the Belgium Government to
11
install solar PV in public health centers, schools and government administration facilities in the
rural areas. [14]
2.3.1 The Barefoot power pack of 5W micro-kit

This micro Solar Home System includes 4lights of 45 lumens each, powered by a 5W solar
panel and stored in a 5Ah lead acid battery. This System will power 4 lights for 12 hours and
can also charge a mobile phone or operate a radio.
Fig. 2.5: The Barefoot power pack of 5W micro-kit used for SHS applications. [5]
2.3.2 BBOXX17 of 50W Solar home system

This system presents a portable battery box of 17Ah that can store up to 200Wh of energy from
a 50W solar panel. In addition to lighting and cell phone charging, these systems can power
small electrical appliances such as a TV, larger ones also fridges, razors etc., in which case an
inverter may be required where the appliance uses an AC current. [5]
Fig. 2.6: BBOXX17 of 50W Solar home system used for rural electrification purposes. [5]
In Rwanda, households are the main consumers of electricity with 51 percent where the larger
portion of electricity is primarily used for lighting purposes whereas the industrial sector with
12
42 percent is the second largest consumer of electric energy, which mainly come from motor-
drivers and lighting. [15]
Fig. 2.7: Main Energy Sources in Rwanda [15].
As shown in the above figure 2.7 which indicates different percentages of total primary energy
consumption in Rwanda, Biomass contributes 85% of primary energy consumed of which wood
contributes a percentage of 57%, Charcoal 23%, Crop residues and peat of 5%. Non-Biomass
sources contribution is 14% of which Petroleum products equal to 11% and electricity
contribution is approximately 4% [15]. Using clean and renewable primary sources for power
generation and reducing diesel-based generation over time along with grid loss minimization
would enable the government to lower the long-term cost of service, to align Generation
capacity and demand and to gradually allow maintained efficient tariff [15].
2.4 Mini-grid Systems

Mini-grids are small distribution systems isolated from the national power system which
includes a source of power generation. This generation is usually provided by hydro or solar,
often in conjunction with either battery storage or diesel generation to handle the intermittency
of power output presented by renewables. It is likely that they could play a role in energy
provision in Rwanda but, given the large investment costs (an EU study for Rwanda estimates
around $1500 per connection – more than grid access), there are several pre-conditions that
need to be met to make them viable. [1]
Increase access to electricity from 17% to at least 60% by 2020 and give access to all schools
and hospitals by 2017 Reduce share of bioenergy in energy demand to 50% by 2020. Expand
the transmission network by 2100 km by 2017. [16]
13
Distance from the grid: Where a connection is available to the national electricity network a
mini-grid is unlikely to be financially viable, since the costs of power generation in a mini-grid
are usually much higher. In deciding where to locate a mini-grid it must be determined whether
this is the least-cost option for providing Tier 3 power to households and businesses. This will,
in a large part, be determined by the proximity of the grid. [1]
Demand: Given the large fixed costs of a mini-grid in both the network and the generation
technology, there needs to be a demand that is both high, and spread across the day for it to be
viable. Most households consume power during the evening but for a mini-grid based on solar
or hydro to be viable it is likely that there would need to be a large demand during the day too
[1].
The government of Rwanda has pledged to set up 100 solar PV mini-grids in rural areas as part
of an effort to mitigate the effects of climate change. Since Rwanda is a country that is very
heavily affected by climate change, the country is taking extensive measures to become a
developed, climate resilient and low carbon economy by 2050. [16] The massive integration of
renewable energy in power systems implies new challenges to the system operator due to their
intermittent nature attributed to climatic conditions. The system operator has limited control
over the amount of electricity produced by these means. Thus, a strong contribution of these
energies can cause imbalances and makes the electric system management more difficult [17].
The inherent intermittence of the renewable energy unit results in fluctuations at the level of the
produced power. Integrating the latter will have a significant impact on the grid. These effects
are manifested by voltage variations and overloads in the network. As far as power quality, both
harmonics and flicker need to be minimized since harmonics result from non-linear electric
loads [17]. Harmonic currents can cause voltage drops that result in the distortion of the supply
voltage. They can also cause resonance in the supply and load components, leading to excess
heat, malfunctions, premature failures and reductions in the lifetime of the transmission and
distribution systems as well as in the electrical components. In the PV system, harmonics are
caused by the conversion from DC to AC. A low-pass filter is normally used at the inverter
output to reduce the harmonics. [17]
For the sake of cost reduction and de-risk investment for prospective mini-grid developers,
Government will first identify eligible sites and undertake a financial and technical feasibility
study. Where the provision of access through a mini-grid represents the least cost option,
Government will undertake measures to stimulate demand, through either policy or investment.
These sites will then be tendered out to private developers. [1] Rwanda’s abundant natural
resources and growing demand for electricity make the development of renewable energy in the
country an investment with large upside potential [16].
Rwanda is a country endowed with plentiful natural resources. Despite its large reserves of
methane gas, large river systems and high levels of solar irradiation, Rwanda’s electric
generation capacity amounts to barely 186 MW, spread out among its 10.5 million residents
[16]. In February 2015, the first utility-scale solar energy project in East Africa was
commissioned at the Agahozo-Shalom Youth Village in Rwanda as shown by the figure 2.8
14
below taken from Gigawatt global, providing 8.5 MW of grid-connected power to 15,000
homes. This increased total grid capacity by 6% [8].
Fig. 2.8: utility-scale of 8.5MW PV power plant constructed in Agahozo-Shalom Youth Village in Rwanda [9].
The above PV Power plant is 20 hectares (49 acres) of land and uses 28,360 photovoltaic panels
and produces 6% of total electrical supply of the country. The project was built with U.S.,
Israeli, Dutch, Norwegian, Finnish and UK funding and expertise [9].
Off-grid renewable energy systems are not only urgently needed to connect a vast number of
people with a source of electricity, but are also most appropriate due to geographical constraints
and costs for grid extension. At the same time, off-grid systems could become an important
vehicle to support the development of renewables-based grids. In developed countries including
Rwanda, mini-grids are increasingly considered an option to improve energy security, power
quality and reliability, as well as to avoid power blackouts due to natural disasters [7].
Despite the growing attention and market opportunities, there are to date only limited data
available and only inadequate definitions of what exactly constitutes an off-grid renewable
energy system. Furthermore, data sources are scarce and inconsistent across countries and
regions. To address this challenge, IRENA has identified many of key areas where
methodological improvements are needed. These methodological improvements include: 1) an
overview of systems; 2) a categorization of off-grid renewable energy systems based on their
application and system design; 3) consistent indicators to differentiate, evaluate, compare and
aggregate data on off-grid renewable energy systems, including hybrid systems; and 4)
measures to compile existing data sources, identify their limitations and create consistency. [7]
15
Table 2.1: Total number of solar power plants and total capacity [9].
No Plant Name Installed Capacity (MW)

1 Jali 0.25
2 Ndera 0.16
3 Gigawatt / Rwamagana 8.5
Total 8.91
Although, only 24.5% of Rwanda’s population has access to electricity, the country’s location
just a few degrees south of the Equator makes it a prime candidate for the development of solar
PV plants but this is not yet implemented in potential areas since most of energy is generated
by Hydro and thermal power plants as illustrated by the figure 2.9 below [16].
Fig. 2.9: Rwanda's total production in terms of energy generated by various sources [16].
16
Fig. 2.10: Rwanda's total energy consumption through public and private services [16].
Rwanda’s total electrical generation capacity amounts to 186 MW, 59% of which comes from
the country’s 7 largest hydroelectric power plants [16]. Thermal power generation, mostly from
diesel and heavy oil fuel, account for the remaining 40% of electrical generation capacity, with
methane gas representing barely 1% of total capacity [16]. Rwanda’s national electrification
rate sits at 24.5%, which 23% coming from on-grid systems and only 1.5% coming from off-
grid-systems [16]. This leaves over several million of the country’s residents without access to
electricity and ranks Rwanda among the countries with the least annual electricity consumption
per capita [16].
Table 2.2: Electricity tariff for residential and non-residential consumption in Rwanda [17].
End-users Electricity tariff in Rwanda (RWF // USD per kWh)

Low Voltage (Industrial) 182 // 0.23
Medium Voltage (All residential and Non-residential) 126 // 0.16
2.5 Overview on HOMER Software

HOMER means Hybrid Optimization of Multiple Energy Resources. It is the main tool used in
this research project and it was developed by the National Renewable Energy Laboratory
(NREL) which is a division the United States Department of Energy. This software is useful in
handling various technologies including PV, boilers, wind, hydro, fuel cells, and loads which
may be AC/DC, thermal and hydrogen. HOMER is an hourly simulator which is used as an
optimization tool for deciding the system configuration. It is used in both developing as well as
developed countries to analyze the off-grid electrification issues. [18]
The approach of HOMER is quite robust and done in three stages to provide a comprehensive
analysis of the entire system as well as a clarification on the technical and financial implications
of design choices. The three stages are simulation, optimization, and sensitivity analysis [19].
17
There is a study conducted by S.M. Shaahid and I. El-Amin [20] to examine solar system for
evaluating the best techno-economic option of hybrid System composed of PV-diesel-battery
to respond to the load requirement regarding the chosen remote village with demand of 15,900
MWh. This research paper has been selected because it includes a solar PV as one renewable
energy source which is related to my project.
HOMER accepts input for the daily load values in hourly increments. This load profile is
extrapolated to the whole year. While it was noted that seasonal changes are not extreme near
the equator to grossly affect the load curve, slight random variability was added to account for
day to day variability, and to make the load more realistic. It has an attractive ability to do the
evaluation of many equipments and resource options over a varying range of constraints with
much effectiveness and high efficiency. [21] These are the main reasons why this software has
been used in a various number of literatures for system design and optimization since it becomes
easy to get a least-cost feasible option as well as optimum sizing of the entire system.
18
3 Data Collection and selected site

PV system design and optimization requires data collection as a preliminary stage for the project
work to be carried out successfully. This include the selection of the site at which the project
will be implemented for future work, load profile analysis for the selected village, Analysis of
solar irradiance in the chosen area and evaluation of renewable resources available taking into
consideration the topography of the region together with geographical characteristics of that
specific location.
3.1 Introduction
Rwanda is divided into five provinces which act as intermediary between the national
government and their constituent districts. The map shows the new provinces of Rwanda which
include: The Southern Province, the Northern Province, the Western Province, the Eastern
Province, and Kigali Province which is a province-level city and the capital of Rwanda.
Fig. 3.1: Rwanda administrative map with a clear illustration of all provinces and their districts. [22]
Rwanda is a country located to the south of the equator and its climate is more moderate. The
tropical climate is influenced by the mountainous landscape and the temperature is stable
throughout the year at approximately 28 °C. [23] The rainy season in Rwanda is from September
through to May, the wettest months are April and November. During the sunny season, the
average daily radiation in the month of July is around 5.09 kWh/m2/d and the driest season is
observed from June to August.
19
3.2 Rwanda Energy Sector Overview

The Government of Rwanda, under its latest Economic Development and Poverty Reduction
Strategy, envisions transitioning from a developing country to a middle-income country by
2020. But Rwanda’s ability to achieve this ambitious goal is constrained by challenges in the
power sector. Thus, Rwanda’s government is working to target 100 percent access to electricity
by 2020. Although Rwanda is endowed with several natural resources, including hydro, solar,
and methane gas, and has plans to generate 563 megawatts (MW) of electric power from these
sources, it currently only has about 186 MW of installed generation capacity to serve a
population of more than 10.5 million people [24].
Fig. 3.2: Rwanda electricity profile showing the current installed capacity and rural access to electricity [24].
Based on current data, Rwanda’s national electrification rate has reached 24.5% which implies
1.5% off-grid and 23% on-grid but more seven million people still lack access to electricity.
Despite high resource potential and opportunities for cross-border export, Rwanda’s power
sector faces significant challenges, including a constrained transmission system. The annual
consumption of electricity per capita is among the lowest in Africa, with approximately half of
consumers using an average of less than 20 kWh per month. [24]
3.3 Data for Bugesera District

Bugesera is a densely populated district in the eastern province of Rwanda with a population
density of 282 people per km2. It covers a total surface area of 1337 Km² and it is composed of
15 Sectors, 72 Cells and 581 Villages with a total Population of 363,339 people, where 177,404
are males and 185,935 are females with an annual Population Average Growth Rate of 3,1%.
[25] This area has no distinct temperature seasons; the temperature is relatively constant during
20
the year. Temperatures drop sharply at night and July is on average the month with high quantity
of solar radiations as the driest season of the year is mostly observed in this month. [26] The
time around July is driest. Rainfall and other precipitation peaks around the month of April as
shown by the Figure 3.3 below.
Fig. 3.3: Monthly average values of various climate data for Bugesera District. [26]
Bugesera District is situated in the South-Eastern plains of Rwanda notably in the south west of
the Eastern Province. It borders are Republic of Burundi (Kirundo Province) in the south,
Ngoma district in the East, Kigali city and Rwamagana district in the North. The district is
sandwiched between Rivers Nyabarongo and Akanyaru which converge at the southern part to
form Akagera River. Bugesera district’s area is characterized by numerous lakes, the biggest of
which are Rweru and Cyohoha. [25]
21
Fig. 3.4: Administrative Map of Bugesera District with all different sectors and rivers. [25]
The topography of Bugesera is characterized with a mixture of plateaus with an altitude varying
between 1,100m and 1,780m and undulating hills dominated by varying heights. Most
prominent of these hills are; Kabuye (1,772m the highest), Juru (1,667m), Maranyundo
(1,614m), and Mwendo (1,575m). The relief is also constituted by a succession of low-plateaus
with hills and dry valleys. The district is equally rich in marshlands alongside rivers; they cover
an estimated area of 6,100 ha and are exploited at an average of 46.3%. [27]
22
Fig. 3.5: Map of Geography allocation of Nyamata Sector in Bugesera district. [28]
Fig. 3.6: View of Kanazi village of Nyamata Sector showing the topography. [29]
23
The largest proportion of Bugesera population is rural based with just a small percentage living
in urban areas as reported by EICV3 statistics proved that, 7% of people in Bugesera live in
town while 93% are rural based. [30] Bugesera district is among the districts with a high
percentage of the population identified as poor since it was ranked to be the 13th country-wide
by percentage of extreme-poor and poor population categories. [30]
The poverty line was defined based on the level of household consumption per adult below
which a household is deemed to be poor. [30]
Fig. 3.7: A Typical house for the poor family in Kanazi cell of Bugesera district. [31]
Even though the energy sector registered great achievement in the last few years, there is still a
long way to go in Bugesera because a negligible number of people have access to electricity
and Electric power is far from arriving in all the significant centres of the District and due to
the current low incomes of the populations, it will be difficult to count on it to satisfy the
requirements in energy. The primary sources of energy used for lighting by households were
categorized as follows: electricity, oil lamp, firewood, candle, lantern, battery, and other
unspecified sources. [30] Due to the problem of accelerated poverty in this area, Bugesera DDP
(District Development Plan) was elaborated within the general frame work of the second
generation of Economic Development and Poverty Reduction Strategy (EDPRS2) whose
overall aim is to increase the quality of life of all Rwandans through rapid and sustainable
economic growth. [27]
3.4 Single house load profile

The selected house for a poor family is composed of few electrical and electronic equipments.
These include 4lamps for lighting purposes, 2 cell-phones as communication facilities and 1
radio serving as a source of daily information. The daily power consumption in watt and daily
24
load profile for this specific single house are presented in the Table 3.1 and in Figure 3.8 below
respectively. The table 3.1 gives the load assumed for a single house as used in this study
(compare to appendix D for more details related to single house load profile).
Table 3.1: Daily load evaluation for a poor single house in Kanazi village.
No. Equipments No. in Power Total Hours/day Watt-

use (W) Power hrs/day
1 Lamps 4 11 44 5 220
2 Cell-Phones 2 5 10 1 10
3 Radio 1 10 10 12 120
4 TV 0 120 0 3 0
5 DVD Player 0 30 0 3 0
6 Computer 0 100 0 2 0
7 Refrigerator 0 500 0 4 0
8 Iron 0 1000 0 1 0
9 Water pumps 0 500 0 1 0
Total 350
No. of houses 100
Total for Poor 35000
families
Based on these values, a typical daily load curve for a single house in Kanazi cell, Nyamata
Sector of Bugesera District with hourly resolution has been described and it is shown in Figure
3.8 below.
The maximum power consumption demand with respect to the obtained load profile, for the
chosen single house is approximately equal to 0.2 kW but with the random variability of 10 %
(standard deviation: daily and hourly noise to make the load data more realistic) for each day
and time step to time step, this maximum demand can become 0.07kW with the energy
consumption of around 0.2 kWh.
25
0.25
0.2
0.15
Load [kW]
0.1
0.05
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
Fig. 3.8: Average daily load profile for a typical single house in Kanazi village selected for case study.
3.5 Solar resources analysis for Nyamata Sector

For sizing a photovoltaic system or any other system that uses solar energy to convert it into
electricity or thermal energy, it is very important to have data relating to solar radiation. [3] The
solar resource information used for selected village at a location at 2o8’ S latitude and 30o5’ E
longitude was derived by PVGIS for Africa. [32] Data on the monthly averages of the daily
radiation sum on a horizontal surface are plotted in Figure 3.7. In addition, tabulated monthly
averaged daily insolation incident are given in Table 3.1, including the calculated clearness
index values for each month. The clearness [5] is defined as a measure of the fraction of the
solar radiation that is transmitted through the atmosphere to the earth's surface.
Fig. 3.9: Monthly global radiation for the selected village in Nyamata, from Homer Software.
26
For an optimal use of photovoltaic energy, a good knowledge of all the data describing the solar
potential of the area is essential [3]. The Large amount of solar radiation for the selected site
can be observed in the month of July since the value is 5.87 kWh/m2/day. The annual average
solar radiation was found to be 5.415 kWh/m2/day and the corresponding average clearness
index calculated by Homer is 0.541 as shown by the Table 3.2 below.
Table 3.2: Monthly global radiation for the selected village in Nyamata sector, from Homer tool.
Month Clearness Index Daily Radiation (kWh/m2/d)

January 0.532 5.47
February 0.531 5.59
March 0.531 5.59
April 0.502 5.07
May 0.527 4.98
June 0.597 5.41
July 0.638 5.87
August 0.586 5.72
September 0.551 5.68
October 0.529 5.53
November 0.485 4.99
December 0.5 5.08
Average 0.541 5.415
27
3.6 Additional resources for the selected site

Additional climatic parameters such as air temperature, earth temperature, wind speed,
atmospheric pressure and relative humidity for the selected site in different months of the year
and their annual average values are presented in the Table 3.3 below.
Table 3.3: Average monthly values for other climatic parameters in Nyamata Sector. [33]
Month Air Relative Wind speed Atmospheric Earth

temperature °C humidity % m/s pressure, kPa temperature, °C
January 25 43.30% 3.8 89.7 26.4
February 25.5 44.50% 3.9 89.7 27.2
March 24.2 62.60% 3.9 89.6 25.4
April 22.8 75.00% 3.7 89.7 23.6
May 22.3 75.50% 3.6 89.8 22.6
June 22.2 68.70% 3.5 89.9 22.4
July 22.4 62.20% 3.5 90 22.7
August 22 69.70% 3.5 90 22.4
September 21.7 75.20% 3.6 89.9 22.1
October 21.7 78.10% 3.7 89.8 22.1
November 22 72.70% 3.6 89.8 22.3
December 23.5 54.80% 3.6 89.8 24.2
Annual 22.9 65.20% 3.7 89.8 23.6
28
4 PV System Components behaviors and Costs

The performance and cost of each component of the PV System are the most important
parameters to consider before the design process. The main components that will be described
in this chapter are the solar panels and the storage battery. There are various types of PV System
configuration based on the kind of voltage system and bus that interconnect the source to the
loads. The two most common types are known as DC coupled system and DC/AC coupled
system. In this energy research project, DC coupled PV system has been chosen because the PV
generating source is connected to a DC bus that is used to supply DC loads present within the
selected house in Kanazi village. These loads include, four lamps for lighting purposes, one
radio as a source of information and two cell phones for communication facilities.
Fig. 4.1: General diagram of the PV system showing the main components. [3]
A complete electricity-management system using PV as its source for power will contain, in
addition to the photovoltaic array, one or more of the following:
 Storage devices
 Power-conditioning equipment, including devices to
- Limit current and voltage in order to maximize power output
- Convert direct-current to alternating current
- Match the converted DC electricity to the utility's AC electrical network
- Safeguard the utility network and its personnel from possible damage caused by
the PV system. [34]
There are two basic forms of electricity: alternating current (AC) and direct current (DC) [34].
The difference between them is that the alternating current (AC) is made up of electrons
alternately flowing in one direction and then in the opposite direction under the influence of a
cycling force (voltage) that acts a part of a time in one and then the opposite direction, while
29
for DC electricity, the electrons flow in a single direction and it is generated by devices such as
batteries and photovoltaic systems [34].
In a battery, electrons gather at an electrode as the result of a chemical reaction within the
battery. In the PV cell, the electrons are generated by light and the ability of the PV cell to move
charge carriers to opposite sides of the cell. The electrons move because there is a driving force,
a voltage which is characteristic of the electric source, for example an electrochemical cell
(battery) or a PV cell [34].
Direct current is a perfectly useful form of electricity for many applications. At an isolated
location, there is no need to do anything more than use PV-generated electricity and perhaps
store it for times when there is no daylight to activate the cells. This is possible so long as the
devices being powered can use direct current. But some types of motors and appliances cannot
be designed for direct current [34]. PV arrays are useful energy producers only when the sun is
shining on them and thus are unproductive a good deal of the time. This extends the availability
of electricity through periods when there is no illumination. If a utility grid is convenient to the
PV system, then it can be used as a low-cost way to store electricity. Excess electricity from the
PV system can be suitably made compatible with grid electricity [34]. When more PV electricity
is being generated than is being used, the excess can be metered and fed to the grid. When the
PV system is not providing enough power, the extra amount needed can be purchased from the
utility grid [34].
4.1 Solar Panels

A PV module consists of several interconnected solar cells (typically 36 connected in series)
encapsulated into a single, long-lasting, stable unit. The key purpose of encapsulating a set of
electrically connected solar cells is to protect them and their interconnecting wires from the
typically harsh environment in which they are used. For example, solar cells, since they are
relatively thin, are prone to mechanical damage unless protected. In addition, the metal grid on
the top surface of the solar cell and the wires interconnecting the individual solar cells may be
corroded by water or water vapor [35]. The two key functions of encapsulation are to prevent
mechanical damage to the solar cells and to prevent water or water vapor from corroding the
electrical contacts.
A fundamental characteristic of a photovoltaic system is that power is produced only while
sunlight is available [36].The current and power output of photovoltaic solar panels are
approximately proportional to the sun’s intensity. At a given intensity, a solar panel's output
current and operating voltage are determined by the characteristics of the load [37]. To generate
large amount of power output, solar cells are connected in a compound structure within the PV
module and the output power from a typical solar cell is approximately equal to 1 watt.
4.1.1 Electrical characteristics of a solar panel

A Solar Panel produces its maximum current when there is no resistance in the circuit, i.e. when
there is a short circuit between its Positive and Negative terminals. This maximum current is
30
known as the Short Circuit Current and is abbreviated as Isc. When the Panel is shorted, the
voltage in the circuit is zero. Conversely, the maximum voltage occurs when there is a break
in the circuit. This is called the Open Circuit Voltage (Voc). Under this condition, the resistance
is infinitely high and there is no current, since the circuit is incomplete. These two extremes in
load resistance, and the whole range of conditions in between them, are depicted on the I-V
Curve. Current, expressed in Amps, is on the vertical Y-axis. Voltage, in Volts, is on the
horizontal X-axis [35].
Fig. 4.2:The I-V and Power Curves for the Solar Panel [38].
A Current (I) versus Voltage (V) Curve of a solar panel shows the possible combinations of its
current and voltage outputs. A typical I-V curve for the solar panel is shown at Figure 2
presented above. The power (P) available from a photovoltaic device at any point along the
curve is just the product of the voltage and the current.
Mathematically,
𝑃 = 𝑉×𝐼 (4.1)
 The short circuit current: Isc corresponds to the short circuit condition when the
impedance is low and is calculated when the voltage equals 0. I (at V=0) = Isc. It occurs at
the beginning of the forward-bias sweep and is the maximum current value in the power
quadrant. For an ideal cell, this maximum current value is the total current produced in the
solar cell by photon excitation. Isc = Imax = Iℓ for forward-bias power quadrant.
 The open circuit voltage (VOC): occurs when there is no current passing through the cell.
V (at I=0) = Voc. It is also the maximum voltage difference across the cell for a forward-
bias sweep in the power quadrant. Voc= Vmax for forward-bias power quadrant
31
At the short circuit current, the power output is zero, since the voltage is zero. At the open
circuit voltage point, the power output is also zero, but this time it is because the current is zero.
There is a point on the knee of the I-V Curve where the maximum power output is located and
this point is called the Maximum Power Point (MPP). The voltage and current at this Maximum
Power Point are designated as Vmp and Imp. The rated power of the solar panel in Watts (Pmax)
is derived from the above values of voltage Vmp and current Imp at this Maximum Power Point
(MPP): Rated power in Watts, Pmax = Vmp x Imp [35].
Example of I-V Curve and Ratings of a 12 V Solar Panel.
Fig. 4.3: I-V Curve and ratings of Solar Panel designed to be connected to a 12V battery [39].
The Maximum Power Point: in the example curve given above is where Vmp is 17 Volts, and
the current Imp is 2.5 amps. Therefore, the rated or the maximum power Wmax in watts is 17
Volts times 2.5 Amps, or 42.5 Watts. The I-V curve is also used to compare the performance of
PV Modules. The curve is, therefore generated based on the performance under Standard Test
Conditions (STC) of sunlight and device temperature of 25 °C. It assumes there is no shading
on the device. Standard sunlight conditions on a clear day are assumed to be 1,000 Watts of
solar energy per square meter (1000 W/m2 or 1 kW/m2). This is sometimes called one sun, or
a peak sun. Less than one sun will reduce the current output of the PV device by a proportional
amount. For example, if only one-half sun (500 W/m2) is available, the amount of output current
is roughly cut in half [39].
 The total current I: in an ideal cell, is equal to the current Iℓ generated by the photoelectric
effect minus the diode current ID, as shown in the equation below:
𝑞𝑉
𝐼 = 𝐼𝑙 − 𝐼𝐷 = 𝐼𝑙 − 𝐼𝑜 (𝑒 𝑘𝑇 − 1) (4.2)
Where,
Io is the saturation current of the diode,
32
q is the elementary charge 1.6x10-19 Coulombs,

k is a constant of value 1.38x10-23J/K,
T is the cell temperature in Kelvin, and V is the measured cell voltage that is either
produced (power quadrant) or applied (voltage bias).
Effect of Series Resistance (RS) and Shunt Resistance RSH

The series resistance is caused by the movement of electrons through the emitter and base of
the solar cell, the contact resistance between the metal contact and the silicon and the resistance
of metal grids at the front and the rear of the solar cell. The shunt resistance is due to
manufacturing defects and lightly by poor solar cell design. It corresponds to an alternate current
path for the photocurrent [40]. The total current in terms of both series and shunt resistance is
given by the following equation:
𝑞(𝑉+𝐼.𝑅𝑠)
𝑉+𝐼.𝑅𝑠
𝐼 = 𝐼𝑙 − 𝐼𝑜 (𝑒𝑥𝑝 𝑛.𝑘.𝑇 − 1) − (4.3)
𝑅𝑆𝐻
Where,
n is the diode ideality factor (typically between 1 and 2), and
RS and RSH represents the series and shunt resistances.
Expanding the equation gives the simplified circuit model shown below. During operation, the
efficiency of solar cells is reduced by the dissipation of power across internal resistances. These
parasitic resistances can be modeled as a parallel shunt resistance (RSH) and series resistance
(RS), as depicted in Figure 4.4 below.
Fig. 4.4: Simplified Equivalent Circuit Model for a Photovoltaic Cell [38].
For an ideal cell, RSH would be infinite and would not provide an alternate path for current to
flow, while Rs would be zero, resulting in no further voltage drop before the load. Decreasing
RSH and increasing Rs will decrease the fill factor (FF) and PMAX as shown in Figure 4.6
33
below. If RSH is decreased too much, VOC will drop, while increasing RS excessively can cause
ISC to drop instead.
Fig. 4.5: Effect of series (a) and shunt(b) resistance on I-V characteristics of the PV Cell [41].
The Fill Factor (FF) is essentially a measure of quality of the solar cell. It is calculated by
comparing the maximum power to the theoretical power (PT) that would be output at both the
open circuit voltage and short circuit current together.
Fig. 4.6: Getting the Fill Factor from the I-V Curve [38].
A larger fill factor is desirable, and corresponds to an I-V sweep that is more square-
like. Typical fill factors range from 0.5 to 0.82. Fill factor is also often represented as a
percentage.
Efficiency is the ratio of the electrical power output Pout, compared to the solar power input, Pin,
into the PV cell. Pout can be taken to be PMAX since the solar cell can be operated up to its
maximum power output to get the maximum efficiency. [37].
𝑃𝑜𝑢𝑡 𝑃𝑀𝐴𝑋
Ƞ= → ƞ𝑀𝐴𝑋 = (4.4)
𝑃𝑖𝑛 𝑃𝑖𝑛
34
Pin is taken as the product of the irradiance of the incident light, measured in W/m2 or in suns
(1000 W/m2), with the surface area of the solar cell [m2]. The maximum efficiency (ƞ𝑀𝐴𝑋 )
found from a light test is not only an indication of the performance of the device under test, but,
like all the I-V parameters, can also be affected by ambient conditions such as temperature and
the intensity and spectrum of the incident light [42].
4.1.2 Effect of Temperature on the performance of a Solar panel

It is important to understand the effect of the temperature on the output performance of the solar
module. Temperature affects how electricity flows through an electrical circuit by changing the
speed at which the electrons travel. The module temperature has a dramatic effect on voltage
parameters. Open circuit voltage and maximum voltage are decrease with increasing module
temperature. This is due to an increase in resistance of the circuit that results from an increase
in temperature. Likewise, resistance decreases with decreasing temperatures [43].
There are many different methods used for increasing the efficiency of the solar module, such
as water cooling methods that are employed to reduce the temperature of the panel and other
methods like dust cleaning, mppt tracking, effective selection of cell materials as well as panel
orientation are also used. [44]
Fig. 4.7: Temperature Effect on I-V Curve [38].
The Performance of solar cells is dependent on environmental conditions and their output
parameters such as output voltage, current, power, and fill factor vary by temperature.
Experimental results showed that the most significant changed by temperature is voltage which
decreases with increasing temperature while output current slightly increase by temperature.
Reduction in the open-circuit voltage for silicon solar cells is about 2mV/°C. As well as the
effect of temperature on the maximum power output is menus 0.005 mw/°C [40].
35
Table 4.1: Output Parameters of Solar Cell Under Different Temperatures [40]
4.1.3 Effect of irradiance on the performance of a Solar panel

The sun radiation plays an important role on the performance of photovoltaic (PV) solar
modules due to its variation from time to time, the output of PV module increases as the sun
radiation increases. The amount of sunlight received by any surface on earth will depend on
several factors including; geographical location, season, local landscape, time of the day and
local weather. The light’s angle of incidence on a given surface will depend on the orientation
since the Earth’s surface is in round-form and the intensity will depend on the distance that the
light should travel to reach the respective surface [45].
Fig. 4.8: The effect of Irradiance on I-V curve of a solar panel [46].
The above Figure 4.8 shows I-V curves for irradiance values of 20, 100, 200 and 600 W/m2.
The I-V curve at 600 W/m2 is for a module that faces the sun directly. When the sun is exactly
overhead in tropical countries, the module should be horizontal for maximum current. When
the sun is low in the sky at 30° above the horizon, the module should be tilted towards the sun
at an angle of 60° from horizontal for maximum current. However, modules are usually fixed
in one position. Thus, they face the sun directly only for a few times a day [46]. Therefore, the
current produced by the module which position is fixed varies through the day, even when the
weather is clear with no clouds. When the sun is covered by thick clouds, the irradiance could
be reduced to 100 W/m2 or less. The reduction in the intensity of irradiation involves a
reduction in the photocurrent because the photocurrent is practically proportional to luminous
36
flow and since the open circuit voltage is also related to the short circuit current, it will undergo
a small reduction [47]. From the above Figure 4.8, the first vertical line represents the empty
battery and the second line represents the fully charged battery for different values of irradiance
and temperature. Therefore, it is optimal to use the 12V battery since it always matches with
changing climatic conditions. For crystalline silicon modules, the main material properties are
temperature coefficient and the solar radiation coefficient, which are normally given by the PV
Manufacture having values of 0.0045K and 0.12 respectively [40]. Considering a solar panel
with variations in ambient temperature and irradiance, the cell temperature (in oC) can be
estimated quite accurately by using the linear approximation which results to the following
useful formula below,
𝑇𝑁𝑂𝐶𝑇 −20
𝑇𝐶 = 𝑇𝑎 + 0.8𝑘𝑊/𝑚 2 ×𝐼(𝑡) (4.5)
Where,
TC is the cell temperature; Ta is the ambient temperature and TNOCT is the nominal
operating cell temperature [48] [4].
4.2 Battery storage
The Energy storage represents a critical part of any energy system, and chemical storage in the
form of a battery is the most frequently employed method for long term storage. In any
photovoltaic system that includes batteries, the batteries become a central component of the
overall system which significantly affect the cost, maintenance requirements, reliability, and
design of the photovoltaic system. Electrochemical battery storage is the most utilized method
for storing electrical power [34]. Because of large impact of batteries in a stand-alone
photovoltaic system, understanding the properties of batteries is critical in understanding the
operation of photovoltaic systems. For systems in which the photovoltaics is the sole generation
source, storage is typically needed since an exact match between available sunlight and the load
is limited to a few types of systems.
By being part of the circuit into which electrical power from the PV supply flows, the battery
keeps the electrical load more nearly constant, and the PV array can be easily designed to
operate more nearly at its optimum power output [34]. Of course, the battery must be protected
from being overcharged by a highly productive array, and a darkened array must be protected
from current flowing to it from a charged battery [34]. This is readily done by placing protective
electronic components such as diodes, which limit the flow of electricity to a single direction
and devices comparable to automobile voltage regulators that prevent battery overcharge while
maintaining a uniform system voltage [34]. Note that a battery storage system without a
regulator must be larger than one so equipped because it must be large enough to ensure against
overcharge [34]. The important battery parameters that affect the photovoltaic system operation
and performance are the battery maintenance requirements, lifetime of the battery, available
power and efficiency [36].
37
Fig. 4.9: Vision 6FM55D, 12V, 55Ah Battery storage [49].
The general features for the above type of battery are:

 Absorbent Glass Mat (AGM) technology for efficient gas recombination of up to 99%
and freedom from electrolyte maintenance or water adding.
 Not restricted for air transport.
 Can be mounted in any orientation.
 Computer designed lead, calcium tin alloy grid for high power density.
 Long service life, float or cyclic applications.
 Maintenance-free operation.
 Low self-discharge.
Fig. 4.10: Surrette battery 6CS25P, 6V. [50]
38
Some distinctive advantages of Surrette battery: [50]

 Premium Battery Brand
 Excellent Stock Availability
 Fast Order Processing & Delivery
 Technical Back Up & System Support
 Competitive Commercial Terms
 Controlled Distribution Channels
For the above battery to last longer, it is important that it is properly charged. Over and under-
charging a Rolls battery will result in shortened service life. The best protection from improper
charging is the use of a quality charger and routinely checking that the charger current and
voltage settings are maintained. [51]
Fig. 4.11: Capacity curve of the Surrette battery 6CS25P, 6V, from Homer.
Fig. 4.12: Life cycle curve for Surrette battery 6CS25P, 6V, from Homer.
39
The capacity curve for the Surrette battery 6CS25P is shown in Figure 4.2. The nominal capacity
of this battery type is indicated as one point on this capacity curve and specified by the
manufacturer as 1156 Ah. The life cycle curve for the Surrette battery 6CS25P is shown in the
above figure 4.3. The Figure indicates clearly that the number of cycles to failure drops rapidly
when the depth of discharge arises and this can be used as a proof to show that the battery
lifetime also depends on the number of cycles to failure.
4.3 Solar Inverter and MPPT System

A solar inverter is needed by the PV system in case of AC loads to convert the generated DC
voltage into AC voltage as required by different AC loads. Also for insuring the best
performance of the PV modules, an MPPT system is important and since the technology has
improved this system is usually incorporated with solar inverters in most stand-alone PV
systems. The efficiency of an inverter varies per load demand and their efficiency curve is
usually provided by the manufacture. An inverter provides a factor of compatibility, but its use
can reduce the array's available electricity if it is not suitably designed to match the electrical
load [34]. Essentially an inverter is a set of automatic switches that provide polarity reversals
from the solar array as shown by the figure below. If there is no utility-grid interconnection,
then the direct-current supply need only be converted to an alternating current output that will
suitably power the AC equipment on hand [34].
Fig. 4.13: Inverter's circuit diagram with different electronic switching circuits [34].
40
In the system illustrated above, when switches 1 and 2 are closed the AC output is positive at
the left terminal. When switches 3 and 4 are closed the positive AC terminal is on the right.
Opening and closing, alternately, the switches in pairs (1/2 and3/4) produces square waveform
as shown in the bottom portion of the illustration [34].
4.4 Other Devices

Most other power-related equipment placed in a PV system is likely to be present because the
PV system is hooked to a utility grid. In such cases, it is important that meters record the
quantities of electricity being sold to an being bought from the utility. Other pieces of PV system
equipment that utilities probably will require are automatic lock-out switches or isolation
transformers which are used for ensuring the separation of the PV and utility-grid systems in
the event of a grid failure [34].
4.5 Cost Estimation for Solar Panels, Battery storage and Inverter
The cost of solar panels is a variable that usually depends on the time, place and size of the solar
panel required for the entire Solar home system installation. The difference between reported
and modeled pricing is due to various factors pertaining to the inherent variability in the type
of systems that are installed each year, their location, state and federal policies, and the ability
of buyers and sellers to agree to a sales price which may be effected by the supply and demand
market factors.
Referring to the reported modeled system prices for different types of solar panels depending
on system size, the recently updated prices are as follows [52].
 Residential Small commercial (≤5 kW) is $ 3.12 /W
 Large commercial (>100 kW) was $ 2.17/W (median)
 Utility-scale (≥8 MW, ground-mounted) was $ 1.78/W (capacity weighted average).
41
Table 4.2: Range of Cost Estimation for monocrystalline Solar Panels depending on the system size [44].
Based on the information given by the above Table 4.2, the approximated cost for the PV system
size between 190W and 560W, used in homer is $1000 [53]. I have chosen different sizes for
Monocrystalline solar panel [54]. The analysis in Homer software is based on the current system
sizes that are used in Rwanda between 30W and 200W. For the storage system, I also used
various types of battery with different nominal voltages and currents to compare the simulation
results and propose the optimal system. The first battery selected in Homer for solar home
system is named Vision 6FM5DD was selected in Homer with a cost of $478 and it has the
following specifications: 12V as nominal voltage, 55Ah as nominal current handling capacity
[55]. The second battery chosen in Homer library for 10kW stand-alone village system is
described as Surrette 6CS25P of 6V, 1156Ah nominal capacity and with a cost of $1200 per
one single unit which implies that for 8 batteries in series the total cost is $9600 [56].
42
5 PV Systems Modelling
The focus in this chapter is the design of a basic solar home system and mini-grid system by the
help of the optimization software named HOMER. The design process begins by enumerating
the important inputs data that demonstrate the technical specifications, resources data and the
costs which are relevant for designing the entire system in HOMER tool.
5.1 Design of 0.2kW Solar Home System

The inputs data used for designing the solar home system in Homer for the selected area are
summarized below.
 PV panel size is 0.2 kW, with a capital cost of $1000, replacement cost of $500 and the
maintenance together with operating cost per year were assumed to be $10.
 Average load demand is 0.008 kW, peak load demand is 0.071 kW, and average of the
daily demand is 0.2 kWh/d and the load factor of 0.117.
 Solar irradiance is 5.41 kWh/m2/d, the clearness index is 0.541 and the average
temperatures is 22°C.
 The battery is Vision 6FM55D of 12V, 55Ah (0.66 kWh) and with a round trip
efficiency of 80 %.
Fig. 5.1: Single line diagram of the basic Solar Home System in Kanazi village.
43
Fig. 5.2: Inputs for modelling a solar home system of a single house in HOMER.
5.1.1 Primary Load Input

The primary load for the single house is the load required to be met immediately in such a way
that there will be no unmet load [56]. This load was measured in different hours of the day for
the selected houses in Kanazi cell to obtain the average daily load profile and the seasonal
profile for a typical single house. Therefore, Homer tool calculates the annual average load, the
peak load and the load factor based on the value entered for each hour of the day with DC load
type.
44
Fig. 5.3: Input window for primary load of a typical single house in Homer.
The baseline data is defined as many of 8,760 values recorded for each hour that are used to
represent the average value of electricity demand and it is expressed in units of Kw
[56].Basically, there are two main technics for production of baseline data such as: either by
using Homer to synthesize data, or by importing hourly data from a selected file [56]. The
difference between these technics is that the one of synthesizing data by Homer just enters at
least one load profile, which is described as a set of different 24 hourly values of electric load
[56]. It is possible to enter different load profiles for different months, and for weekdays and
weekends too. But if only one load profile has been entered, it will be used throughout the year.
5.1.2 PV inputs
In this project work a monocrystalline solar panel type SUNTECH STP250S is selected and
related specifications are mentioned in the data sheet and on the name plate by the manufacturer
[54]. In using Homer software, the selected PV system type is supposed to work under
maximum power point to achieve the desired output power form this system. The important
parameters required by homer for the PV input are summarized in the Figure 5.4 below.
45
Fig. 5.4: Homer Input window for PV system used for electricity generation for a house in Kanazi cell.
The above PV input window is used to show the cost characteristic curve for PV panels, select
the range of sizes that can be evaluated by Homer to obtain the optimal system for single house
analysis and indicated the position of the solar panel. The cost of PV panel depends mainly on
the system size as shown by the cost table in the above Figure 5.3. Since PV costs are generally
assumed to be linear with size, there is a necessity of using only a single row [56].During the
development of this work, the capital investment cost of the selected PV panel size of 0.2 kW
has been specified and considered to be $1000 and the replacement cost is assumed to be $500.
The operation and maintenance (O&M) cost is taken as $10.
5.1.3 Battery inputs

The input window displayed in Figure 5.5 below, contains all the types battery that are present
within the homer component library. When the Homer user needs which type of battery to be
considered, Homer provides many various types of batteries and their different properties by
just clicking on details button. It is then easier to choose an appropriate battery type from the
list based on predetermined specification that can efficiently match with the used PV system
production. After selecting the optimal battery type, the user must enter the chosen values for
capital cost, replacement cost, operation and maintenance cost and the size to consider for the
46
storage system depends on the number of batteries per string regarding the desired nominal
voltage and current or the optimal configuration.
Fig. 5.5: Input window for the battery selected for solar home system design in Homer.
In this project work, the type of battery selected is Vision 6FM55D having a nominal voltage
of 12V and 55Ah as nominal current capacity. For solar home system configuration, one battery
is enough as shown in the above Figure 5.4 but for a bigger system if the number of batteries is
k, then the costs for the total has be multiplied respectively. Finally, different costs for one
battery unit have been considered as follows $478 for initial capital investment, $250 for
replacement and $ 5 for annual O&M [55].
5.1.4 Modelling of Solar resource inputs

A resource in Homer terminology is defined as any external parameter that can be used by a
component for production of electricity or heat [56]. In this research project, the solar radiation
input has been taken as the main resource needed for optimum performance of the selected
photovoltaic system configuration. The entered data for average daily radiation for each month
for the village was provided by PVGIS (Photovoltaic Geographical Information System) for
Africa. The latitude and longitude are used to specify the exact location of selected village site
on the Earth's surface depending on the specified time zone. These two parameters are very
47
important for obtaining the average daily radiation for each month that will enable homer to
calculate the monthly clearness index of the chosen area. In this research work, the selected
village in Nyamata used as case study for PV system is located at 2o 13’ latitude South and 3o
6’ longitude East as displayed by the Figure 5.6 below.
Fig. 5.6: Resource window with data for Kanazi village taken from PVGIS entered [19].
The Yearly scaled data for solar radiation intensity in Kanazi village based on hourly load for
each day of every month is shown in Figure 5.7 below. Therefore, Homer uses the obtained
PVGIS radiation data for Africa to compute the hourly solar radiation data based on the monthly
average radiation and clearness index for the selected area as shown in appendix B for more
details.
48
Fig. 5.7: Yearly scaled data for solar radiation intensity in Kanazi village based on hourly load.
Based on the results from the above Figure5.7, It is seen that for the selected village the sun
rises around 7 in the morning with irradiance varying between 0.14 to 0.42 kW/m2 and sets at
6 in the evening. The peak hours of sunshine are observed from 9 am to 3 pm with irradiance
variation in the range of 0.56 to 1.26 42 kW/m2 and the average number of peak hours is
assumed to be equal to 5 hours per day.
49
5.2 Design of 10kW off-grid PV System for the village

The following are the inputs data used for designing the stand-alone village PV system and the
design was done using Homer software so that both results will be later compared, discussed
and analyzed in the next chapter.
 PV panel size is 10 kW, with a capital cost of $35000, replacement cost of $25000 and
the maintenance together with operating cost per year were assumed to be $30.
 The peak load demand is 3.8 kW, and average of the daily demand is 48 kWh/d with a
load factor of 0.528.
 Solar irradiance is 5.41 kWh/m2/d, the clearness index is 0.541 and the average
temperatures is 22°C.
 The selected battery for this project work is 6CS25P-Surrette of 6V presenting a nominal
capacity of 1,156 Ah (6.94 kWh) and a round-trip efficiency of 80% with the capital
cost of $ 1200, then the approximated replacement cost and O&M costs for one unit of
this battery have been, assumed as $ 1200 and $ 30/year, respectively.
Fig. 5.8: Single line diagram for a village PV system of 10kW capacity in kanazi cell.
50
Fig. 5.9: Inputs for modelling a 10kW PV system for kanazi village using Homer Software.
The main factors that have been considered in this design include the solar resource that
indicates the clearness index as well as the average daily radiation for each month based on
latitude and longitude, the PV panel, the average primary load, the battery type, the converter
type, the AC and DC buses. Therefore, Homer uses the solar resource inputs to calculate the PV
array power for each hour of the year and it also uses the latitude value to calculate the average
daily radiation from the clearness index and vice-versa. From The above figure 5.9, the load
demand for this village can increase up to 48kWh/day but in this case study, we will consider
residential and commercial loads that can be supplied by a 10kW PV system.
5.2.1 Battery input

For obtaining the number of batteries to be used for optimal configuration, the user of Homer
software can use the name plate shown in the figure below and first choose which type of battery
to be used depending on the needed storage capacity as well as its cost [56].
51
Fig. 5.10: Battery storage window with properties in homer component library.
The above library contains different types of battery found in homer component library and this
provides the possibility for the Homer user to select from the existing list which type of battery
to be used for system design and modelling. By clicking on details button on the right-hand side
of the box, Homer displays many more various kind of battery storages and their different
properties. For this case study, the chosen battery type is Surrette 6CS25P of 6V and 48V
voltage system that requires 8 strings composed of 8 batteries in series.
As indicated by the below formula, the battery charging efficiency is assumed by the software
to be the square root of the round-trip efficiency [56]:
𝜂𝑏𝑎𝑡𝑡,𝑐 = √𝜂𝑏𝑎𝑡𝑡,𝑟𝑡 (5.1)
𝜂𝑏𝑎𝑡𝑡,𝑑 = √𝜂𝑏𝑎𝑡𝑡,𝑟𝑡 (5.2)
Where
𝜂𝑏𝑎𝑡𝑡,𝑐 : charge efficiency,
𝜂𝑏𝑎𝑡𝑡,𝑑 : discharge efficiency
𝜂𝑏𝑎𝑡𝑡,𝑟𝑡 : round-trip efficiency [56]
52
Fig. 5.11:Battery input window, from homer showing properties and cost curve.
The above figure 5.11 represents the window into which can be found all the necessary details
regarding the battery selected for this case study. Those details show properties and cost
estimation including the capital cost of $1200, replacement cost of $1200 as well as annual
operation and maintenance costs of $30 [56]. The total cost will depend on the number of
batteries multiplied by the cost per one single battery unit [56].
5.2.2 Converter input

A converter is needed for any system that is composed of both the AC and DC configuration
[56]. It can be an inverter in case of DC to AC conversion or a rectifier for AC to DC or can
serve both purposes. In this case, we have considered it as an inverter since it is used to convert
the DC power generated by PV system into AC power supplied to loads.
53
Fig. 5.12: Converter input window, from homer showing the cost curve and different sizes for the converter.
The above figure 5.12 shows the details concerning the sizes to be considered for optimal
system, different types of costs as well as the cost curve for the converter referring to the capital
and replacement costs. The cost of inverter is related to its size and in this case the capital as
well as replacement costs for 10 kW converter is considered at $ 11,370. The efficiency of 90%
used by the inverter for conversion from DC into AC form is provided by the manufacturer of
the inverter and it is entered as input data for further Homer calculations, and the rectifier
efficiency for converting AC into DC form is entered as 85% [56]. The project life time is
defined as the number of operational years for the entire system [56]. In this case study, the life
time for the selected inverter was estimated to be 15 years from Homer as displayed in the above
figure 5.12.
5.2.3 Economic inputs

Homer software is useful for obtaining the optimal system configuration based on the Levelized
COE analysis by calculating the Net Present Value (NPV) including all the costs for the lifetime
of the project [56]. Furthermore, Homer classifies different types of configuration depending
on how incremental is the total NPC and LCOE values [56].
54
Table 5.1: Economic inputs used by Homer for calculating the NPC for the system.
The total Net Present Cost (NPC) of the system is the difference between the present values
of all the costs that occur over the project lifetime and the present values of all the revenue
earned over project lifetime [56]. The formula below is used to calculate the net present value
of the costs that will be made with n year later.
1+ i′
𝐶𝑁𝑃𝐶 = 𝐶( 1+𝑑 )n (5.3)
Where
𝑖′: is the annual inflation rate (%)
d: is the nominal interest rate (%)
The interest rate is used to convert between one-time costs and annualized costs [56]. It is also
known as discount rate. Since renewable energy technologies such as solar PV systems have a
higher capital cost but with lower operating and maintenance cost, the Levelized COE analysis
must be conducted for the system depending on the life time to facilitate the economic
evaluation of stand-alone systems supplied by solar PV systems. In contrast, due to fuel costs,
generator maintenance and replacement costs fossil fuel based electricity generation systems
have lower capital cost but higher operation and maintenance cost compared to the existing
renewable energy technologies. Therefore, LCOE analysis can be used to compare the
economic viability of various technological solutions for cost effectiveness.
55
6 Results
This chapter will be focusing on the analysis of results obtained from Homer simulations and
the comparison between solar home systems and off-grid village systems based on the cost and
availability. The simulation results are essential in obtaining the optimal PV system which
means a system that is available and can supply electrical power at a lower cost referring to the
total NPC, Levelized COE as well as the cost of operation.
6.1 Results for Solar Home System
Fig. 6.1: Simulation results based on cash flow summary and Net Present Cost categorized by component.
The results obtained from different PV system sizes have been analyzed based on the Net
present cost and the overall performance of the system using the same battery with different PV
production capacity and the maximum size for SHS used in Rwanda for DC loads is 200W.
This battery was chosen since it has the required nominal capacity for voltage and current to be
used for solar home systems in Rwanda.
56
6.1.1 Overall Optimization Results

The solar home system designed for a typical single house with DC loads in Kanazi village was
composed by the PV production capacity of 0.2 kW and a battery storage system of 55Ah as
nominal capacity with a nominal voltage of 12V. Since the mentioned nominal capacity
corresponds to 0.66 kWh, this implies that the used storage system can supply power to different
loads in the house for a period of 3 days in case of low radiation intensity at the selected site.
Table 6.1: Overall Optimization Results without shortage.
From homer optimization results in the above Table 6.1, representing a screen shot displaying
the summary of outcomes for different size of PV system used for simulation considering no
shortage. These include the initial capital, operating cost per year, total net present cost, capacity
shortage for each system and the renewable fraction that explains the availability of the system
as well as increase in reliability of supply from lower to bigger size. Therefore, since the
renewable fraction was found to be the same from 0.03 kW to 0.2 kW this gives the possibility
of designing a bigger PV system for the entire village community. The chosen optimal system
configuration for my case study is the PV system size of 0.2kW, 1 Vision 6FM55D battery and
the power dispatch strategy will depend on the load demand.
57
Table 6.2: Overall Optimization Results with Shortage.
The above table 6.2 displays the optimization results considering a maximum annual capacity
shortage of 50% and a minimum renewable fraction of 10%. The lowest energy cost is observed
for a solar home system of 80W which is $1.780/kWh and for a SHS of 200W capacity, the
energy cost is $2.614/kWh. Therefore, it is more economical to use one SHS of 200W instead
of using two systems of 80W.
6.1.2 Sensitivity Results

By considering zero shortage and zero renewable fraction, the optimum system size for solar
home system calculated by Homer was found to be 120W capacity with an initial capital of
$1.078, operating cost of 61$/year, total NPC of $1.783 and the COE in $/kWh is 2.130.
58
Table 6.3: Tabular Sensitivity Results without shortage.
59
Table 6.4: Tabular Sensitivity Results with shortage.
60
Fig. 6.2: Levelized cost of energy vs PV array capacity with a fixed renewable fraction of 10%.
In the above figure 6.2, the red line represents the Levelized cost of energy and the yellow line
represents the PV array capacity regarding the maximum capacity shortage for the system.
Considering a case of the maximum capacity shortage of 50% and minimum renewable fraction
of 10%, when the capacity shortage is less than 40% the LCOE and the PV array capacity are
directly proportional but above 40% they become inverse. Therefore, the optimum size for solar
home system calculated by Homer is 30W presenting an initial capital of $678, operating cost
of $27 per year, total NPC of $984 and the COE is $2.298/kWh as shown by the above table
6.4. This implies that the reliability of supply depends upon the maximum capacity shortage
and the minimum renewable fraction of the system.
61
Fig. 6.3: Graphical Sensitivity Results based on Levelized COE.
The above figure 6.3 displays the graphical sensitivity results calculated by Homer based on the
Levelized COE which is defined as the cost per kWh of electrical energy consumed throughout
the lifetime of the PV system. The LCOE analysis has been performed considering a lifetime
of 25 years with all relevant costs namely the initial capital investment and operating as well as
maintenance costs. The Levelized COE for electrical power produced by a photovoltaic system
can be calculated from the following equation below.
€
𝑇𝑜𝑡𝑎𝑙 𝐴𝑛𝑛𝑢𝑎𝑙𝑖𝑧𝑒𝑑 𝐶𝑜𝑠𝑡 (𝑦𝑒𝑎𝑟)
𝐿𝐶𝑂𝐸 = 𝑘𝑊ℎ (6.1)
𝑇𝑜𝑡𝑎𝑙 𝐸𝑛𝑒𝑟𝑔𝑦 𝐶𝑜𝑛𝑠𝑢𝑚𝑒𝑑 (𝑦𝑒𝑎𝑟)
62
6.2 Results for 10kW Village PV System
The analysis was done for different PV sizes using Homer software and a specific system of
10kW capacity was chosen to be optimum since it is more economical and reliable in terms of
cost and availability of the supply compared to the use of individual solar home system. In
addition, the key input parameters have been modified for sensitivity analysis to be performed.
Fig. 6.4: Simulation results based on cash flow summary and Net Present Cost categorized by component
The above figure 6.4 shows the selected PV system of 10kW capacity with an inverter of 10kW
and 8 batteries per string. The total NPC calculated by Homer is equivalent to $63,685 and the
Levelized COE as well as the cost of operation are $0.494 per kWh and $673 per year
respectively. The system architecture obtained displays the simulation results in details based
on cash flow summary and the NPC categorized by each component considering the capital,
replacement, salvage, operation and maintenance costs.
63
Table 6.5: Overall Optimization Results with no Shortage.
Table 6.6: Categorized Optimization Results with zero shortage
The above tables 6.5 and 6.6 show homer optimization results, representing a screen shot
displaying the summary of outcomes for different sizes of PV system used for simulation
considering no shortage. These include the initial capital, operating cost per year, total net
present cost, capacity shortage for each system and the renewable fraction that explains the
64
availability of the system as well as increase in reliability of supply from lower to bigger size.
Therefore, since the renewable fraction was found to be the same from 20kW to 30kW this
gives the possibility to select a 20 kW PV system for the entire village community in case of
increase on demand side management.
Table 6.7: Overall Optimization Results with Capacity shortage
Table 6.8: Categorized Optimization Results with capacity shortage
65
The above optimization results are not enough to be considered for obtaining the optimum
sizing of the PV system since Homer does not cover all costs related to the performance of the
village system. Also, the inputs data entered in Homer may not be fully accurate because they
were taken from PVGIS and this may result in a small difference between those inputs and the
real data. Therefore, the above optimization results can be made more reasonable by considering
some uncertainties in the input variables and this can be achieved through the sensitivity
analysis.
Optimal System Type System Types

50
0.726 0.587 0.537 0.534 0.494 0.494 PV/Battery
Superimposed
Levelized COE ($/kWh)
40
0.726 0.587 0.537 0.534 0.494 0.494
Min. Ren. Fraction (%)
30
0.726 0.587 0.537 0.534 0.494 0.494
20
0.726 0.587 0.537 0.534 0.494 0.494
10
0.726 0.587 0.537 0.534 0.494 0.494
0.726 0.587 0.537 0.534 0.494 0.494

0
0 10 20 30 40 50
Max. Annual Capacity Shortage (%)
Fig. 6.5: Graphical Sensitivity Results based on LCOE:
The optimal system type present a Levelized COE of 0.587 considering the sensitivity analysis
which is different from the LCOE obtained by optimization results which was 0.494 but the
difference is very small about 0.093 and this is due to the intermittency of renewable energy
resources which varies depending on different climatic conditions making them unpredictable.
66
Table 6.9: Tabular Sensitivity Results with shortage.
67
Fig. 6.6: LCOE vs PV array capacity with a fixed renewable fraction of 10% for the village system of 10kW.
The above table 6.9 displays the sensitivity results considering a maximum capacity shortage
of 50% and a minimum renewable fraction of 10%. Also, the figure 6.6 above shows the LCOE
by a red line and the PV Array capacity by a yellow line which implies that they are proportional
to each other and this provides a chance for designing bigger systems. In this case the optimum
sizing for PV system is 10kW with 8 Surrette batteries and a 10kW inverter as highlighted in a
blue color in the table above, all having an initial capital of $55,970, operating cost of $673 per
year, total NPC of $63,685 and the COE of $0.494 per kWh. Sensitivity analysis also shows in
table 6.9 that in terms of reliability and economic constraints, it is much more efficient to use
two systems of 10kW each instead of using a 20kW system
68
7 Discussion
The aim of this master’s thesis was to analyze the use of PV system to convert solar energy into
electric power to supply the electricity need in rural areas. The preliminary design step was
based on the comparison of different solar home systems for a typical single house in Kanazi
village, Nyamata sector of Bugesera district. This village is located at a far distance from
national grid lines and most of the population do not have access to electricity from the grid.
The Bugesera district receives a large amount of sunlight with an annual average solar radiation
of 5,28 kWh/m2/day. In addition, Bugesera is one of the driest site among other districts of the
eastern province of Rwanda since its average temperature is high, with values above 21°C and
the precipitation amount in this area is the lowest in the whole country, with values below 900
millimeters per year. Therefore, the above factors have been taken into consideration while
selecting Bugesera district as an interesting site on which this study was performed, using solar
PV technology as a renewable energy resource for electricity. A fixed daily load profile shown
in Figure 3.3 has been assumed throughout the year because Rwanda is located closer to the
equator, therefore Bugesera district is not affected by seasonal variations and the day length
does not change in a significant way. The average daily load demand for the single house in
Kanazi village is 200Wh/day and the peak load is 71W, thus the system size has been chosen
to be above that peak of load.
The used overall system consists of a PV panel with a rated power of 200W and a storage battery
consist of Vision 6FM55D with a nominal capacity of 55 Ah, 12V. For this system, the
maximum annual capacity shortage calculated by homer is 50% and the minimum renewable
faction is 10%. Further analysis was performed based on different PV sizes used in Rwanda for
existing solar home system (SHS) in the range of 30 W to 200W, to identify which system is
more efficient and compare it with a 10kW PV village system. If we compare in terms of total
cost for the village with only SHS to a village system of 10kW It is evident that the village
system becomes more economical. For example, considering a village of 300 homes having
each a SHS of 200W that costs $1478 including additional cost of operation and maintenance
means that the total is 300×$1478 which equals $443400 yet for the village system of 10kW
that can supply a minimum of 50 houses the cost is $55970 which means that it is required to
have 6 systems of 10kW each to supply 300 houses and the total cost for those systems is
$55970×6 which equals $335820. Therefore, there is a difference of $107580 that can be saved
by taking an option of using a much bigger solar PV system for the rural community with a
suitable operation and maintenance scheme which can ensure the sustainable operation of the
system.
Moreover, the obtained solar home system solution and village system can only be implemented
in other rural locations if the environment’s conditions such as solar radiation, sun hours and
temperature requirement are similar. Furthermore, the PV system performance and efficiency
should also remain unchanged but there is a possibility of moving to the design of bigger off-
grid system in case of high load demand. A study conducted for a year by Zakaria Bouzid and
Nassera Ghellai to ensure an optimal configuration and to prevent energy deficit has proven that
69
for many hours of the year, the batteries are fully charged and the energy produced by the
photovoltaic panels is lost. Therefore, an increase in the total energy storage capacity could
solve this problem for a better management of energy flow. [3] Instead of waiting for the grid
power in case of increase in demand, the best option is the design of higher capacity stand-alone
PV system of 10kW with a big battery storage capacity that can supply the entire village
community. This is proposed as the best option because the cost of extending the grid to rural
villages is extremely high, around 21.000 $ per km of one power distribution line and the current
national grid electric power price for the domestic consumers in Rwanda lies in the range of 0.2
– 2.4 $/kWh. Therefore, it is confirmed that the cost of the energy consumption is high
compared to the alternative use of photovoltaic system.
70
8 Conclusion and Future work

The main purpose of this master’s thesis project work was to design and compare a solar home
system with an off-grid village system of 10kW capacity to obtain an economical option for
rural electrification in Kanazi cell, Bugesera District of Rwanda which is composed of 300
houses taken for this case study. The goal was achieved by considering the global horizontal
radiation at the selected site and by considering the primary load obtained during the survey
conducted for single house analysis in daily hours considering the cost for an individual solar
home system and the overall cost for the village system. In addition, an evaluation of the climate
data as well as the potential of renewable energy resources available in that area has been done
for finding feasible solutions for the entire population.
In fact, the method of gathering climate data and solar resources has been a useful approach to
understand where this system can mostly be suitable for usage. As shown by the research,
Bugesera district is a region with high intensity of solar radiation and low rain-fed, especially
in Nyamata Sector. Therefore, the use of photovoltaic technologies for electricity production
and supply has been a sustainable solution for the population in rural villages that are far from
national electric grid. Photovoltaic systems are not capable of ensuring a continuous electric
power supply unless if they are connected to an additional storage system. Since this became a
serious challenge during rainy days it was impossible for the PV panels to provide the needed
electrical power. Hence batteries have been associated with solar home system, village system
and other solar based technologies.
Furthermore, after selecting the appropriate components for the entire system and describing
their behavior, both systems have been modelled in HOMER software, and system simulations
have been conducted to determine the best system which can be more economical and reliable
for supplying power to rural areas predicting the increase in load demand and consumption. The
simulation results have shown that it is more advantageous to use a 10kW off-grid system for
village electrification, although the initial capital investment is higher, it is much more
economical throughout the lifetime compared to the grid connection since the electricity prices
in Rwanda stand around $0.226/kWh as announced in the report published by the African
Development Bank regarding the energy sector review and action plan [57]. Solar PV is indeed
a good alternative for electricity generation because it is both cheaper and beneficial towards
energy independence, in addition to being one of the most abundant renewable energy resource
that is available in Rwanda.
Finally, this project work provides a basis and framework for the evaluation of this solution not
only in Bugesera, but also in other parts of the country where the sun shine is relatively intensive
and based on the results from simulation, an additional step of grid connection during the rainy
season can also be embraced. Therefore, the village system of 10kW is selected to be the best
since solar home systems are limited to household applications and the availability of supply is
low. The option for grid connection will depend on the increase of load demand and climate
change and this will enable the solar system to sell power to the grid in case of excess production
of PV modules and to buy power from the grid during the rainy season.
71
9 References
[1] Ministry of Infrastructure, "Rural Electrification Strategy," Republic of Rwanda, June 2016.
[Online]. Available: http://www.mininfra.gov.rw/fileadmin/user_upload/aircraft. [Accessed 17
January 2017].
[2] F. Habyarimana and H. G. Beyer, "Investigating the applicability of photovoltaic solar energy
technologies in rural and urban electrification in Rwanda," in 29th European Photovoltaic Solar
Energy Conference and Exhibition, 2013.
[3] Z. Bouzid and N. Ghellai, "Estimation of Solar Radiation on Inclined Surface and Design Method
for an Autonomous Photovoltaic System. Application to Algeria," Research Unit on Materials
and Renewable Energies, vol. I, no. 9, pp. 1-6, 2015.
[4] Dr. M.Narendra Kumar, Dr. H.S. Saini, Dr.K.S.R. Anjaneyulu and Mr.Kuldip Singh, "Solar
Power Analysis Based on Light Intensity," in International Conference on Innovations in
Electrical & Electronics Engineering (ICIEEE-2014), Hyderabad, 2014.
[5] D. Disch and J. Bronckaers, "An analysis of the off-grid lighting market in Rwanda: sales,
distribution and marketing," GVEP International, Africa Regional Office, London, pp.12-46,
July, 2012.
[6] E. F. Camacho, T. Samad, M. G. Sanz and I. Hiskens, "Control for Renewable Energy and Smart
Grids," Impact of Control Technology, California, 2011.
[7] R. Kempener, O. L. d’Ortigue, D. Saygin, J. Skeer, S. Vinci and D. Gielen, "off-grid renewable
energy systems: status and methodological issues," Paper, p. 5–36, 2015.
[8] "Rwanda renewable energy resources," [Online]. Available: http://www.africa-eu-

renewables.org/market-information/rwanda/renewable-energy-potential. . [Accessed 19 January
2017].
[9] Republic of Rwanda, "Energy Sector Strategic Plan," Ministry of infrastructure, Kigali, 2015.
[10] P. Mahendran and T. Anderson, "Evaluation of a Facade Integrated Concentrating Solar Collector
System," in Solar World Congress, Daegu, Korea, 2015.
[11] S. L. Do and J. C. Baltazar , "Potential energy savings from a semi-transparent solar cell window,"
in Solar World Congress, Daegu, Korea, 2015.
[12] "Online Eletrical Engineering Study Site," [Online]. Available:

https://www.electrical4u.com/components-of-a-solar-electric-generation-system. [Accessed 22
March 2017].
[13] Mobisol, "sites/default/files/Documents/csconnections2_5-4-15_final_lowrez.pdf," [Online].

Available: http://www.uncdf.org. [Accessed 21 October 2016].
72
[14] "Solar Home System: Market Analysis for Rwanda," [Online]. Available:
https://energypedia.info/wiki/Market_for_Solar_Home_Systems_(SHS)_in_East_Africa#Target
_Market_Analysis_for_Rwanda. [Accessed 25 January 2017].
[15] J. d. D. Uwisengeyimana, A. Teke and T. Ibrikci, "Current Overview of Renewable Energy

Resources in Rwanda," Journal of Energy and Natural Resources, vol. 5, no. 6, pp. 92-97, 2016.
[16] "facts and figures about solar energy," Solar Plaza, Nairobi, 2016.
[17] A. Ameur, K. Loudiyi, A. Khallaayoun and M. Aggour, "Dynamic Modelling of a Solar PV

Plant," Renewable Energies Grid Integration Department, vol. 1, no. 9, pp. 1-5, 2015.
[18] R. Sen and S. C. Bhattacharyya, "Renewable Energy-Based Mini-Grid for Rural Electrification:
Case Study of an Indian Village," 2014.
[19] T. Lambert, P. Gilman and P. Lilienthal, "Micro power System Modeling with Homer,"
Integration of Alternative Sources of Energy, pp. 379-418, 2006.
[20] . S. M. Shaahid and I. E.-. Amin, "Techno-economic evaluation of off-grid hybrid photovoltaic-
diesel-battery power systems for rural electrification in Saudi Arabia-A way forward for
sustainable development," Renew. Sustain. Energy Rev, vol. 13, p. 625–633, 2009.
[21] T. Givler and P. Lilienthal, "Using HOMER software, NREL’s micro power optimization model,
to explore the role of gen-sets in small solar power systems," Case Study: Sri Lanka, p. 1–31,
2005.
[22] R. p. m. pdf. [Online]. Available: http://www.mapsofworld.com/rwanda/rwanda-political-

map.html. . [Accessed 19 09 2016].
[23] Bugesera District, "Africa/Rwanda/Bugesera/2301117," [Online]. Available:

http://www.meteovista.com. [Accessed 7 September 2016].
[24] Power Africa , "Rwanda Energy Sector Overview," Power Africa in Rwanda, Kigali, 2016.
[25] "Bugesera District Potentials," [Online]. Available:

http://www.bugesera.gov.rw/fileadmin/templates/document/BUGESERA_DISTRICT_POTEN
TIALS.pdf. [Accessed 20 January 2017].
[26] Bugesera District,

"travel/pax/q/384476/Bugesera/RW/Rwanda/0/#1|0|travel|q|0|384476|Bugesera|RW|Rwanda|Af
rica," [Online]. Available: http://www.chinci.com. [Accessed 12 September 2016].
[27] "Bugesera District Development Plan (2013-2018)," [Online]. Available:

http://www.bugesera.gov.rw/fileadmin/user_upload/Bugesera_District_DDP_2012-
17_Final.pdf. [Accessed 5 February 2017].
[28] Eastern Rwanda, "Nyamata Map," [Online]. Available: https://cse.google.com. [Accessed 5

November 2016].
73
[29] Eastern Rwanda. [Online]. Available: https://cse.google.com. [Accessed 26 January 2017].
[30] "Integrated Household Living Conditions Survey (EICV3 report)," Kigali, 2011.
[31] Bugesera District Housing, "wiki/File:Rwanda_Bugesera_district_Housing.jpg," [Online].

Available: https://commons.wikimedia.org. [Accessed 14 September 2016].
[32] PVGIS, "Solar resource information," [Online]. Available: http://re.jrc.ec.europa.eu. [Accessed

14 February 2017].
[33] RET Screen Data, "NASA Surface meteorology and Solar Energy," [Online]. Available:
https://eosweb.larc.nasa.gov. [Accessed 7 November 2016].
[34] Solar Energy Research Institute, Basic Photovoltaic principles and methods, Colorado: Technical
Information Office, 1982, pp. 1-67.
[35] "Modules Structure," [Online]. Available:

http://www.pveducation.org/pvcdrom/modules/modules-structure. [Accessed 09 February
2017].
[36] Batteries overview. [Online]. Available: http://www.pveducation.org. [Accessed 22 January

2017].
[37] "Solar Panel Efficiency," [Online]. Available: http://www.wholesalesolar.com/solar-

information/solar-panel-efficiency. [Accessed 27 January 2017].
[38] E. A. Daman, A. J. Boukar and S. . A. Almayri, "Investigation and Performance Evaluation of

Solar Cell," in 3rd International Conference on Automation, Control, Engineering and Computer
Science (ACECS-2016), Hammamet-Tunisia, 2016.
[39] Learning Center, "Solar Panels Characteristics," [Online]. Available:

http://www.samlexsolar.com/learning-center/solar-panels-characteristics.aspx. [Accessed 30
January 2017].
[40] D. M. Tobnaghi, R. Madatov and d. naderi, "The Effect of Temperature on Electrical Parameters
of Solar Cells," pp. 1-4, December 2013.
[41] A. Mcevoy; , T. Markvart and L. Castaner, "Solar cells: materials, manufacture and operation,"
2004.
[42] "Physics of Solar Cells," [Online]. Available:

http://depts.washington.edu/cmditr/modules/opv/physics_of_solar_cells.html. [Accessed 3
February 2017].
[43] I. B. Karki, "Effect of Temperature on the I-V Characteristics of a Polycrystalline Solar Cell,"
pp. 2-6, August 2015.
[44] A. Pradhan and M. Ali, "Irradiance and Temperature effect on Solar PV Performance," Advanced
Electrical and Computer Engineering (IJAECE), vol. 3, no. 1, pp. 1-3, 7 April 2016.
74
[45] M. S. Salim, J. M. Najim and S. M. Salih, "Practical Evaluation of Solar Irradiance Effect on PV
Performance," pp. 2-5, November 2013.
[46] "Effect of irradiance on the performance of a Solar panel," [Online]. Available:

http://www.solarfasttrack.com/English/chap5/chap5d.htm. [Accessed 13 February 2017].
[47] F. Bouzid and S. B. Machich, "The effect of solar spectral irradiance and temperature on the
electrical characteristics of a ZnO-SiO2-Si (N) photovoltaic structure," pp. 1-12, June 2010.
[48] John Wiley & Sons Ltd, Handbook of Photovoltaic Scienceand Engineering, A. Luque and S.
Hegedus, Eds., Chichester, Atrium: Laserwords Private Limited, 2003, pp. 296-297.
[49] "Battery types: Vision 6FM55D," [Online]. Available:

http://yhipower.com.au/data/vision/fmd_series/6FM55D.pdf. [Accessed 29 January 2017].
[50] "World Class Battery Engineering," [Online]. Available: http://www.off-grid-

europe.com/downloads/dl/file/id/40/rolls_battery_brochure.pdf. [Accessed 27 March 2017].
[51] "Rolls Battery," [Online]. Available:

http://www.energistar.com/documentation/Rolls%20Battery%20User%20Manual.pdf.
[Accessed 28 March 2017].
[52] D. Feldman, G. Barbose, T. James, S. Weaver, R. Fu and C. Davidson, "Photovoltaic System

Pricing Trends 2015 Edition," 2015.
[53] "Approximate cost of PV panels," [Online]. Available: http://www.wholesalesolar.com/solar-

information/offgrid-ballpark-cost. [Accessed 7 February 2017].
[54] "Monocrystalline Solar Panel SUNTECH STP250S-20/Wd - Wind Turbine Solar Kit Solar
Panels batteries," [Online]. Available:
http://www.merkasol.com/epages/62387086.sf/en_GB/?ObjectPath=/Shops/62387086/Products
/panel018. [Accessed 10 February 2017].
[55] "Vision 6FM55D: Deep Cycle Battery," [Online]. Available: http:// http://www.vision-batt.com/
batteryguru.com.au/Vision-VRLA-6FM55D-12V-55Ah-Industrial-Battery-AGM-Deep-Cycle-
Battery. [Accessed 24 January 2017].
[56] "Homer," [Online]. Available: http://www.slideshare.net/WWP19910307/homer-47932711.

[Accessed 28 January 2017].
[57] African Development Bank, "Energy Sector Review and Action Plan - Rwanda," 2013.
[58] Notton. G, Cristofari. C, Mattei. M and Poggi. P, Modelling of a double-glass photovoltaic

module using finite differences, Appl. Therm.Eng, 25, ed., New York: Springer international
publishing, 2005, p. 2854–2877.
75
Appendices
Appendix A Mobisol customers in Kanazi cell, Nyamata
sector, Bugesera district.
Db id Name Location System size Approve

d at
9837 Rwanda Online/ Bihari, Kanazi, Solar Home System 2016-09-
5 Bihari Nyamata, Bugesera, 200W RW 26
RW
4841 Karanganwa Musagara, Kanazi, Family V8 SHS 100 W 2016-01-
0 Papias Nyamata, Bugesera, 07
RW
4817 Nsanzumuhire Rugando, Kanazi, Solar Home System 2016-01-
6 Celestin Nyamata, Bugesera, 200W RW 05
RW
4406 Shyaka Bernard Musagara, Kanazi, Family V8 SHS 100 W 2015-12-
9 Nyamata, Bugesera, 02
RW
3926 Hagenimana Rugando, Kanazi, Family V8 SHS 100 W 2015-10-
7 Theoneste Nyamata, Bugesera, 22
RW
3275 Nyamata Teacher Musagara, Kanazi, Solar Home System 2015-08-
9 Training Nyamata, Bugesera, 200W RW 24
College(ttc) RW
2346 Barawigirira Rugando, Kanazi, Solar Home System 30W 2015-05-
1 Angelique Nyamata, Bugesera, RW 19
RW
1584 Bagirigomwa Nyarugati I, Kanazi, Solar Home System 2014-12-
5 Octave Nyamata, Bugesera, 120W - V7 -17" 19
RW
1122 Mutesi Rugando, Kanazi, Solar Home System 2014-09-
2 Annonciathe Nyamata, Bugesera, 100W RW 25
RW
6025 Ntakirutimana Nyarugati I, Kanazi, Solar Home System 80W 2014-07-
Marie Rose Nyamata, Bugesera, - V7 - 17" 03
RW
5345 Senzira Jean Rugando, Kanazi, Solar Home System 80W 2014-06-
Bosco Nyamata, Bugesera, - V7 - 17" 15
RW
5308 Bihezande Nyarugati I, Kanazi, Solar Home System 2014-06-
Aimable Nyamata, Bugesera, 120W - V7 -17" 14
RW
76
5063 Nsanzimana Rugando, Kanazi, Solar Home System 30W 2014-06-

Bruno Nyamata, Bugesera, - V7 08
RW
4473 Ngayaberura Nyarugati I, Kanazi, Solar Home System 30W 2014-05-
Gerard Nyamata, Bugesera, - V7 23
RW
4299 Ngoga Emmanuel Rugando, Kanazi, Solar Home System 2014-05-
Nyamata, Bugesera, 120W - V7 -17" 16
RW
4176 Uwamariya Gitovu, Kanazi, Solar Home System 80W 2014-05-
Angele Nyamata, Bugesera, - V7 - 17" 11
RW
3984 Habirora Herman Rugando, Kanazi, Solar Home System 30W 2014-05-
Nyamata, Bugesera, - V7 04
RW
3523 Ukobukeye Rugando, Kanazi, Solar Home System 2014-04-
Philbert Nyamata, Bugesera, 120W - V7 -17" 25
RW
3427 Ndayisaba Daniel Nyarugati I, Kanazi, Solar Home System 2014-04-
Nyamata, Bugesera, 120W - V7 -17" 21
RW
Appendix B Photovoltaic Geographical Information for Nyamata

Sector.
Fig. B1: Interactive map of Photovoltaic Geographical Information System.
77
Fig. B2: PVGIS Estimates of long-term monthly averages for solar irradiation.
Fig. B1: The scaled data on average monthly radiation in 24hrs of different days.
The above Figure B1 displays the scaled data of average monthly radiation on a daily profile of
different days and the peak intensity of solar radiation in the selected village is observed in the
78
month of July which is known as the hottest month but the lowest intensity of sun shine is
recorded for the month of November which is the coldest month of the year.
Appendix C Details on daily consumption for Kanazi village.
79
80
81
82
Appendix D Hourly power distribution per day for selected

houses.
83

Jeannine Uwibambe PDF

Uploaded by

Copyright:

Available Formats

Jeannine Uwibambe PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jeannine Uwibambe PDF

Uploaded by

Copyright:

Available Formats

Design of Photovoltaic System for Rural

University of Agder, 2017

To my beloved Son Prince Daniel IGANZE

3.4 Single house load profile ........................................................................................... 24

Appendix C Details on daily consumption for Kanazi village. ......................................... 79

1.1 Background and Motivation

1.2 Problem Statement

1.3 Goals and Objectives

1.4 Research Methods

1.5 Key Assumptions and Limitations

The final designed system has the following limitations.

2.1 Integrated Renewable Energies in Rwanda

2.1.1 Hydro Power generation

2.1.2 Geothermal energy resources

2.1.3 Wind power

2.1.4 Methane Gas

2.1.5 Peat to power

2.1.6 Biogas power

2.1.7 Solar energy

2.2 Basic operation mechanism of the photovoltaic cell

2.3 Solar Home Systems

2.3.1 The Barefoot power pack of 5W micro-kit

2.3.2 BBOXX17 of 50W Solar home system

Fig. 2.7: Main Energy Sources in Rwanda [15].

2.4 Mini-grid Systems

No Plant Name Installed Capacity (MW)

End-users Electricity tariff in Rwanda (RWF // USD per kWh)

2.5 Overview on HOMER Software

3 Data Collection and selected site

3.2 Rwanda Energy Sector Overview

3.3 Data for Bugesera District

3.4 Single house load profile

No. Equipments No. in Power Total Hours/day Watt-

3.5 Solar resources analysis for Nyamata Sector

Month Clearness Index Daily Radiation (kWh/m2/d)

3.6 Additional resources for the selected site

Month Air Relative Wind speed Atmospheric Earth

4 PV System Components behaviors and Costs

4.1 Solar Panels

4.1.1 Electrical characteristics of a solar panel

q is the elementary charge 1.6x10-19 Coulombs,

Effect of Series Resistance (RS) and Shunt Resistance RSH

4.1.2 Effect of Temperature on the performance of a Solar panel

Fig. 4.7: Temperature Effect on I-V Curve [38].

4.1.3 Effect of irradiance on the performance of a Solar panel

Fig. 4.9: Vision 6FM55D, 12V, 55Ah Battery storage [49].

The general features for the above type of battery are:

Fig. 4.10: Surrette battery 6CS25P, 6V. [50]

Some distinctive advantages of Surrette battery: [50]

4.3 Solar Inverter and MPPT System

4.4 Other Devices

5.1 Design of 0.2kW Solar Home System

5.1.1 Primary Load Input

5.1.3 Battery inputs

5.1.4 Modelling of Solar resource inputs

5.2 Design of 10kW off-grid PV System for the village

5.2.1 Battery input

𝜂𝑏𝑎𝑡𝑡,𝑐 = √𝜂𝑏𝑎𝑡𝑡,𝑟𝑡 (5.1)

𝜂𝑏𝑎𝑡𝑡,𝑑 = √𝜂𝑏𝑎𝑡𝑡,𝑟𝑡 (5.2)