Analysis of Power System Options for Rural
Analysis of Power System Options for Rural
Electrification in Rwanda
by
Odax Ugirimbabazi
Supervisor: Professor Hans Georg Beyer
Master Thesis in Spring 2015
This Master's Thesis is carried out as part of the education at the University of Agder and is
therefore approved as a part of this education. However, this does not imply that the University
answers for the methods that are used or conclusions drawn.
Faculty of Engineering and Science
University of Agder
Grimstad, 25 May 2015
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Abstract
The development of modernized energy system for developing countries especially in rural areas is
constantly a considerable problem to energy utilities. The progressive use of diesel generators in
rural areas as main source of electrification is continuously becoming unsuitable because of the
following reasons; the diesel generator requires the fuel at every single second of operation and
the maintenance of every time is needed and it is very important to worry about the instability of
power generated by those generators and the accessibility of fossil fuels is still a challenge for some
communities. Whereas the introduction of new technologies by using Renewable Energy systems
RESs has given a hope, confidence and security in electrification of rural communities. With a
combination of RETs, a traditional diesel generation and batteries, a mini power system of the
combination is adequate to manage harmony in operation, therefore granting a stable means of
developing electrical power system to the developing countries especially those ones in rural areas.
The target of this development is the analysis of a mini hybrid power system options to come up with the
best techno-economic and optimum configuration of RETs for supplying electricity to one village in
Rwanda. In this development, a hybrid system with a low cost of energy is presented for
electrification of one of isolated village of Burera district, in Northern Province of Rwanda. First
of all, the renewable resources are determined, an assessment of the predicted village energy
demand is estimated, and using the software called HOMER, a best hybrid system types is
described, elements measured, and the optimization of the system configuration is done to come up
with the reliable and efficient operation in order to answer to the village demand with an
economical cost.
The system type is discovered as follows; a micro hydropower plant, diesel generator and a
compound of batteries and this is found as the best option. In detail, for the case studied the best
hybrid system has the following configuration: a micro hydro power plant (MHPP) of 20 kW, the
diesel generator of 10 kW and the battery bank of 55.5 kWh. The MHPP generates 99.6 % of the
total output, which is approximately 198,000 kWh/yr. The diesel generator is used to supply only
0.4 % of the total generation, resulting in 207 hours of operation annually. The obtained system
configuration has a rough cost of energy of 0.2 $/kWh and may be further reduced to 0.13 $/kWh,
if state subsidies become available for covering 40 – 50 % of the capital investment. It clear that
this hybrid system is more economically viable whether it is operated as off-grid or grid connected.
Keywords: Rural electrification, Renewable Energy, Hybrid System, Power System, Homer, PV
and Hydro.
i
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Preface
This thesis is presented to the Faculty of Engineering and Science, University of Agder, in partial
fulfilment of the requirements for gradation to Master of Science in Renewable Energy. The thesis’
main objective was to explore the techno-economic power system solution which is a renewable
energy-based technology for electrification of one selected village in Rwanda. The work described
here has been conducted under the supervision of Professor Hans Georg Beyer and Programme
coordinator Dr. Stein Bergsmark.
My sincere gratitude goes to my supervisor, Professor Hans Georg Beyer for his great
encouragement, ideas, comments and continuous support throughout the process of project
accomplishment. My special thanks also go to Stein Bergsmark for providing valuable guidance
when writing this thesis. His comments and suggestions have helped me to improve my writing.
Last but not least, my special thanks to Professor Maurice Ghislain Isabwe for his support and
advice throughout my stay at Agder University, to my colleagues who helped me in numerous ways
to make this thesis a success.
Odax Ugirimbabazi
University of Agder
Grimstad, Norway
June 2015
ii
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Contents
Abstract ...................................................................................................................................... i
Preface ....................................................................................................................................... ii
Contents .................................................................................................................................... iii
List of Figures ........................................................................................................................... v
List of Tables ........................................................................................................................... vii
List of Abbreviations .............................................................................................................. viii
1
2
3
4
5
Introduction ...................................................................................................................... 1
1.1
Background and Motivation ....................................................................................... 1
1.2
Problem Statement ...................................................................................................... 1
1.3
Goal and Objectives .................................................................................................... 2
1.4
Literature Review ....................................................................................................... 3
1.5
Research Method ........................................................................................................ 4
1.6
Key Assumptions and Limitations.............................................................................. 5
1.7
Analysis Framework ................................................................................................... 6
1.8
Thesis Outline ............................................................................................................. 8
Data Collection ................................................................................................................ 9
2.1
Introduction................................................................................................................. 9
2.2
Village Load Profile ................................................................................................. 11
2.3
Solar Resource Assessment ...................................................................................... 12
2.4
Hydro Resource Assessment .................................................................................... 15
Hybrid System Components Characteristics and Costs ................................................. 19
3.1
Introduction............................................................................................................... 19
3.2
PV Panels .................................................................................................................. 20
3.3
Micro-Hydro Power Plant......................................................................................... 27
3.4
Diesel Generator ....................................................................................................... 33
3.5
Storage Battery ......................................................................................................... 36
3.6
Inverter ...................................................................................................................... 38
Hybrid System Modelling .............................................................................................. 39
4.1
Introduction............................................................................................................... 39
4.2
Modelling of Equipment ........................................................................................... 40
4.3
Modelling of Resources ............................................................................................ 51
4.4
Modelling of Other Important Factor ....................................................................... 53
Results ............................................................................................................................ 58
iii
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
5.1
Optimization Results ................................................................................................ 58
5.2
Sensitivity Results .................................................................................................... 62
5.3
Futures Connection of the Hybrid System to the National Grid .............................. 66
5.4
Design of the Hybrid System .................................................................................... 67
5.5
Economic Viability ................................................................................................... 69
5.6
Efficient Use of Electricity in the Micro grid ........................................................... 70
5.7
Comparison of Electricity Prices .............................................................................. 70
6
Discussion ...................................................................................................................... 72
7
Conclusion ..................................................................................................................... 74
Appendices .............................................................................................................................. 80
iv
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
List of Figures
Figure 2.1: Map of Burera District .................................................................................................... 9
Figure 2.2 : Map of Geography allocation of Karegamazi site. ...................................................... 10
Figure 2.3 : Closer or zoomed view of Karegamazi village ............................................................ 10
Figure 2.4 : Village load profile ...................................................................................................... 12
Figure 2.5 : Monthly radiation sums for the selected village, from Homer. ................................... 13
Figure 2.6 : Placement of Rugezi catchment in Burera District ...................................................... 16
Figure 2.7 : Reservoir of karegamazi at which the hydropower plant is possible .......................... 16
Figure 2.8 : Discovered and simulated daily stream flow ............................................................... 17
Figure 2.9 : Average monthly stream flow at Rusumo gauging station .......................................... 17
Figure 3.1 : AC coupled hybrid system ........................................................................................... 20
Figure 3.2 : The I-V and Power aspect of a perfect solar cell ......................................................... 21
Figure 3.3 : The equivalent circuit of non-ideal solar with components in dotted line................... 22
Figure 3.4 : The I-V characteristic of PV in the two diode model. ................................................. 22
Figure 3.5 : The effect of resistance on the I-V characteristic of PV .............................................. 22
Figure 3.6 : The dark I-V characteristic of PV in the two diode and series resistance. .................. 23
Figure 3.7 : Effect of solar irradiance and cell temperature on the I–V curve ................................ 23
Figure 3.8 : Solar PV ground mounted system ............................................................................... 27
Figure 3.9 : Micro hydropower plant overview .............................................................................. 28
Figure 3.10 : Diversion Weir and Intake ......................................................................................... 28
Figure 3.11 : Settling Basin ............................................................................................................. 29
Figure 3.12 : Headrace .................................................................................................................... 29
Figure 3.13 : Head Tank .................................................................................................................. 30
Figure 3.14 : The penstock .............................................................................................................. 30
Figure 3.15 : Connection arrangement between Turbine and Generator ........................................ 30
Figure 3.16 : Typical system losses for a system running at full design flow ................................ 31
Figure 3.17 : Typical generator efficiency curve ............................................................................ 34
Figure 3.18 : Capacity curve of the Surrette 6CS25P, 6V battery, from Homer. ........................... 37
Figure 3.19 : Lifetime curve of the Surrette 6CS25P, 6V battery, from Homer. ............................ 37
Figure 4.1 : Inputs required by HOMER hybrid model. ................................................................. 40
Figure 4.2 : Random variability (daily and hourly noise) set to zero. ............................................. 41
Figure 4.3 : Load plot without any added noise for the first week. ................................................. 41
Figure 4.4 : Load plot with an added random variability for the first week. .................................. 42
Figure 4.5 : Homer primary load input window. ............................................................................. 43
Figure 4.6 : PV input window, from homer. ................................................................................... 45
Figure 4.7 : Hydro input window, from homer. .............................................................................. 47
Figure 4.8 : Hydro input window, from homer. .............................................................................. 48
Figure 4.9 : Batteries stored in homer component library. .............................................................. 48
Figure 4.10 : Battery input window, from homer............................................................................ 49
Figure 4.11 : Battery input window, from homer............................................................................ 50
Figure 4.12 : Synthetic solar radiation data over a period of a year. ............................................... 51
v
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 4.13 : Solar resource inputs window, from Homer. ............................................................. 52
Figure 4.14 : Hydro resource inputs window, from Homer. ........................................................... 53
Figure 4.15 : Values of elements optimization................................................................................ 54
Figure 4.16 : Changes in the real interest rate in Rwanda over the past 32 years ........................... 55
Figure 4.17 : Economic input window. ........................................................................................... 56
Figure 5.1 : Summary of HOMER optimization results in categorized way. ................................. 59
Figure 5.2 : Electricity production from the best system type. ....................................................... 59
Figure 5.3 : Optimization results when using only renewable resources. ....................................... 60
Figure 5.4 : Cost flow summary by cost type.................................................................................. 60
Figure 5.5 : Nominal cash flow of the project throughout 20 years. ............................................... 61
Figure 5.6 : Breakeven grid extension distance with its cost .......................................................... 62
Figure 5.7 : HOMER optimization and sensitivity results in categorized way ............................... 63
Figure 5.8 : Surface plot of cost of electricity from hybrid system. ................................................ 64
Figure 5.9 : Line graph for total NPC vs. design flow rate and breakeven grid extension distance64
Figure 5.10 : Number of batteries vs the water flow rate. ............................................................... 65
Figure 5.11 : Converter capacity with respect to the water flow rate. ............................................ 65
Figure 5.12 : Breakeven grid extension distance with respect to hybrid system ............................ 65
Figure 5.13 : LCOE at different design flow rate............................................................................ 66
Figure 5.14 : LCOE at different diesel price. .................................................................................. 66
Figure 5.15 : Single line diagram of the hybrid system .................................................................. 68
vi
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
List of Tables
Table 2.1 : Assumptions on daily consumption for the selected community. ................................. 12
Table 2.2 : Monthly average daily irradiance incident on a horizontal surface for the target location.
......................................................................................................................................................... 14
Table 2.3 : Monthly average daily irradiance on a horizontal surface for Germany....................... 14
Table 2.4 : Monthly mean values for other climatic parameters in Burera District. ....................... 15
Table 3.1 : Items to make a trial calculate of construction cost. ..................................................... 32
Table 3.2 : Approximate Diesel Fuel Consumption Chart. ............................................................. 34
Table 3.3 : Regular and typical diesel maintenance schedule and their estimated costs. ............... 35
Table 3.4 : Cost of Diesel generator on the market. ........................................................................ 36
Table 3.5 : Inverter specifications. .................................................................................................. 38
Table 4.1 : The summary of the costs of components and other relevant costs. ............................. 57
Table 5.1 : Optimal least cost hybrid system for the case study. .................................................... 59
Table 5.2 : Cost summary of the project based on the used component. ........................................ 61
Table 5.3 : Effect of subsidies on the electricity price. ................................................................... 69
Table 5.4 : Effect of system fixed O & M cost on the electricity price. ......................................... 70
vii
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
List of Abbreviations
CC
Cycle Charging
DC
Direct Current
DG
Diesel Generator
DG
Distributed Generation
DVD
Digital Video Disc
EDL
Economical Distance Limit
EICV3
Third Integrated Household Living Conditions Survey
HOMER
Hybrid Optimization Model for Electric Renewables
IPP
Independent Power Producer
LCOE
Levelized Cost of Energy
LF
Load Following
LUCE
Levelised Unit Cost of Electricity
MHPP
Micro Hydro Power Plant
MPPT
Maximum Power Point Tracker
NPC
Net Present Cost
NPV
Net Present Value
PV
Photovoltaic
PWM
Pulse Width Modulation
REG
Rwanda Energy Group
REMA
Rwanda Environment Management Authority
RES
Renewable Energy Sources
RET
Renewable Energy Technology
SHP
Small Hydropower
USA
United States of America
viii
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
1 Introduction
Electricity is the backbone and imperative condition for a country to be developed in terms of
economy and the good quality in terms of lifestyle for the citizens [1]. The estimation shows that
in many developing countries several billion of people do not have mandatory and vital public
services because of not having electricity [1]. In most cases, the extension of electricity is either
impossible because of geographic allocation, or because of high financial involved in the extension
or not enough for the demand. Due to that, the adoption of an off-grid stand-alone RES constitute a
useful option for electricity inadequacies in rural area of the developing countries in which the
evolution in national grid extension continue to be slower than the population growth [2].
1.1 Background and Motivation
The situation of not having enough electricity especially in the rural villages, this is one important
fact that negatively affect the lifestyle of most of Rwandan. The government of Rwanda face the
crisis of granting electrical power to its citizens. Currently, the grid connected is estimated around
23%, where the percentage of rural communities is only 5%. This is although 85 % of Rwandan
live in rural villages, and mainly employ in subsistence farming for nourishment and a means of
securing the necessities of life. In view of Rwanda with a considerable number of populations in
rural area, this introduces the energy sectors and regulators to a number of confrontation in energy
extension and development.
First of all, there is presently inadequate electrical power to satisfy the power demand in Rwanda.
The power production is centralize in the cities or in the developed centers. Furthermore, the cost
for the grid extension combined with the complication of the land in the high hills and mountains
of Rwanda, all of the latter reasons affect the grid expansion with high rate.
High cost of electricity also results to unaffordability of electrical power for rural consumers. This
is connected with their disinclination to contribute for the extension requirement. Thus, the
obligation for government involvement.
Due to these factors the task of extending the grid to the people in order to have access to electricity
is not easy in Rwanda. Instead the village residents are pushed to move to places with existing grid
connection. All these factors have persuaded me to find out the more reliable and sustainable option
for the power production in the rural electrification in Rwanda.
1.2 Problem Statement
The republic of Rwanda has an ambitious target of providing electricity to everyone. In the so called
vision 2020, this will help in transforming the country into middle income economy, where the
goods export will be more than goods import. This is one of the strategic plan for the reduction of
poverty so that the country could end up with the development in its economy [3]. To achieve these
1
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
targets, the involvement of every one is very important. Different way of participation can be used,
research is one way of point out some weak aspect and forecast for the fulfillment of the targets.
Currently no more research have been done for the proper option of renewable systems for rural
energy purposes in Rwanda. Currently, in rural areas most of the schools, health centers,
administration posts and other home house communities use solar systems for each home and fuel
generators.
Instead of providing isolated solar systems for each home or fuel generator, the utilization of RET
for electrification to the whole community in rural villages is more economical and reliable because
the battery capacity of these solar home systems (around 30-100 Wp) is very small. Therefore
during the seasons of low solar radiation, particularly in rainy seasons these systems are not able to
meet the load, so these systems are not 100 % available. This micro grid can be energized by using
renewable energy based on the hybrid system technology, into which multiple combinations of
RETs can be integrated. Furthermore, a kind of dispatching for conventional technology can be
utilized to improve the quality and availability of the service. No matter how, to make the system
economically viable, the appropriate technologies should be attentively privileged and the complex
must be conveniently determined so as to reduce the overall cost [1][4].
In various developing countries, many based hybrid systems projects have been implemented for
rural electrification. Anyway, still a lot of researches are being conducted for the viability and
reliability of using hybrid system for rural electrification projects in various rural communities
around the world; That is why, the same technologies should be established in Rwanda, since the
combinations of RETs in this country is not taken into account, even if there has been a large
improvement in the renewable industry in the past years. Therefore, this project analyses different
combinations of RETs in order to obtain the more techno-economics hybrid system based micro
grid for supplying electricity to a rural community in Burera District in Rwanda.
The Burera district is one of non-electrified districts in Rwanda and it is far from the urban areas.
The EICV3 (Third Integrated Household Living Conditions Survey) results show that the total
population of Burera district in 2010–2011 was 354,000. This means 18% of the total comminity
of Northern Province and 3.3% of the total society of Rwanda [5]. In the Burera district, only 3.2%
of households use electricity as their main source of lighting, this make the district to be the third
ranked after Musanze (14.5%), Gicumbi (8.9) in Northern Province [5]. The blackouts of every day
is also problem for the ones connected to the national grid.
1.3 Goal and Objectives
The aim of this development is to come up with a hybrid power system solution from the best
combination of RET (Renewable Energy Technology) that will use the resources which are
available in Rwandan rural area to fulfill the electricity demand in a reliable, affordable and
sustainable manner with a cost-effective solution.
The achievement of the upper goal, the ability and the accomplishment of the below objectives is
required:


Estimating the everyday load demand of the selected area.
Studying the potential of RE resources in the preferred locality.
2
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda








Describing the relevant renewable energy resources for the proposed hybrid system
The selection of component and the analysis of its cost.
Model electricity produced based on RETs.
Modeling and simulation of the system with the application of HOMER software.
Optimization and sensitivity testing of the system type in HOMER.
Selecting the best option based on the COE (Cost of Energy) generation.
Performance evaluation of the optimal hybrid system.
Compare the optimal hybrid system to the grid extension in terms of costs.
1.4 Literature Review
The optimal design of a hybrid system in terms of cost and the reliability has become of great
importance with the increase in usage of hybrid renewable energy systems. A lot of studies and
researches are being conducted all the day in order to close the knowledge gap that advocates the
requirement for the projects in this regard and to grant support for the method. Numerous researches
accomplished in this field in few decades, especially in remote area electrification but few of them
has been selected in this project because they have some special ideas related to this research[6].
Off-Grid Electrification
Arash Asrari, Abolfazl Ghasemi, and Mohammad Hossein Javidi [7] in their research aims, firstly,
was to explore how to expand the contribution of RES by combining the diesel power sources
and renewable energy sources so that the system can supply electricity to the rural centers in
economical way. On their second stage, they have tried to connect RESs to the national utility grid
in order to realize a more cost effective and techno-optimum system. The software called HOMER
has been used to see the practicability of possible combination of hybrid configuration using dieselRES and distributed power system with RES. The results demonstrate that the RES integration is
a key for cost effective for the system which is certainly cleaner and more climate-friendly [1].
This paper has been selected because, it deal with some technics used for distributed power system
and the combination of renewable energy technology of socio-economic optimization.
Tshering Dorji, Tania Urmee and Philip Jennings [8] in their study the aims was to identify the
least-cost and optimum technologies be used in the rural environment [1]. Their study focuses on
the energy needed by rural communities, resources available to the selected rural area, and policies
and programs that should be fulfilled for the electrification of rural areas. The software HOMER
has been used in hybrid optimization model for the design of distributed generation (DG) systems.
This paper has been selected due to its comparison between the costs obtained from the RETs
systems and the grid extension cost.
Studies on HOMER
HOMER is an acronym which mean Hybrid Optimization Model for Electric Renewables. It is
software developed by the American National Laboratory for Renewable Energy. It can be used for
handling a number of technologies including PV, boilers, wind, hydro, fuel cells, and loads which
3
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
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 [9].
D.Saheb-Koussa, M.Haddadi and M.Belhamel [2] in their study, they deal with the design of hybrid
system. Techno-economic optimization of two renewable sources; photovoltaic and wind, with the
diesel and battery storage has been obtained. Their target was to find the suitable stand-alone hybrid
system that will provide the energy autonomy of remote area with minimum COE. This paper has
been selected, because of having the same target as the one that I have in my project.
E.M. Nfah and J.M. Ngundam [10] who studied a hybrid which including the Pico-hydro and
incorporating a biogas generator. This research has been selected because it use a hydropower as
one renewable energy source.
S.M. Shaahid and I. El-Amin [11] the aim of their study was to examine solar system in order to
evaluate the best techno-economic of hybrid RES composed with PV–diesel–battery to answer to
the load required by the selected remote village with the demand of 15,900 MWh.
Several other literatures have used the Homer software for techno-economic optimum sizing of
hybrid systems. Homer algorithms help in the evaluation of techno-economic feasibility of RET
options and to see the technology with cost effective. It has also integrated with a product database
with different products from a variety of manufactures. Hence this software is widely used for
hybrid system optimization.
Knowledge Gap
The above review shows the popularity of HOMER as a tool to analyze decentralized electricity
supply systems. However, most of the researches do not account electricity demand in rural areas
carefully. As the optimal system configuration is obtained to meet the demand, demand analysis do
an important role. Most of the researches also focus on a limited level of supply and do not often
acknowledge the productive utilization of electricity. Furthermore, whereas technology selections
are based on local conditions, it is likely to investigate alternative combinations more imaginatively.
Finally, studies also limit their scope to techno-economic reasoning and ignore the business issues
or practical considerations related to their implementation [1]. Without such considerations, most
of the development remain theoretical in nature [1]. This chapter tries to bridge the above
knowledge gaps and presents an application of HOMER to extend the scope of the work and
knowledge base [9].
1.5 Research Method
The research will start with data collection of renewable energy resources, establishment of village
load profile, overview of component characteristics and costs, research on hybrid system
configurations, modeling and simulation of the hybrid system, selection of optimum system based
on simulation results and the performance assessment of the selected system.
4
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
First of all, it is necessary to determine the daily load profile of the village. There is no variations
of the load profile due to season changes because due to the equatorial location there are no distinct
summer or winter seasons in Rwanda.
Here, the calculation of the load profile of the village is done via self-performed survey that I could
perform due to my familiarity with this region. In addition, I will use the results from survey forms
for households grid connected which have been conducted on other rural villages connected to
national grid one year ago. I will use parameters such as, the number of households and public
utilities, family income, predisposition and readiness to purchase electrical appliances and potential
small businesses that can emerge with the availability of electricity. These in all is quiet enough for
load demand for the village [1]. However, a reasonable assumption can be used in case where to
get the data from site survey is not possible in order to estimate the load curve. I will use the micro
grid optimization software called HOMER. The simulations are needed to make a considerable
number of hybrid system arrangement that grant several combinations of renewable energy
resources. The lifetime net present cost of the hybrid systems that can supply the village load with
the required level of availability should be calculated to determine the lowest energy cost hybrid
configuration. The sensitivity analysis of the anxieties regarding the system inputs like solar
radiation should be assessed to inspect the best system that can supply the load at the lowest energy
cost for diverse conditions.
1.6 Key Assumptions and Limitations
The scope of this development is limited to determining the best techno-economic combination of
RE resources in a hybrid configuration for electrification of one community selected in Rwanda and
the evaluation for performance of the system is included but this will not deal with the complete
configuration of the micro grid powered by this hybrid system. The analysis of this hybrid system
will be done by considering the following assumptions.






Meteorology and solar energy data from NASA Surface Meteorology and Solar Energy
website represented by RET Screen International are considered to accurate for computing
solar PV systems for off-grid electrification systems[12].
The same annual variations of solar radiation occur all over the project lifetime.
The consumers live conforming to a daily routine coming from the same load cycles every
day, since there is no summer or winter for the selected location because the temperatures
seems to constant in the year.
Rate of inflation will be considered the same for all types of costs (fuel cost, maintenance
cost, labor cost .etc.) occurring all over the 20 years [4].
The hybrid configuration is not location specific and will be the optimal configuration for
other locations where the renewable energy potential is the same as the selected region.
This is a good example for other location in Rwanda, depending with the load profile and
availability of the renewable energy resources. The same approach can be used for other
communities in Rwanda by following the same procedures as it is used throughout this
project.
This study will not discuss the issues related to the micro grid stability and control.
5
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
The designed system will have the following limitations.


Only solar and hydro energy will be chosen for the analysis due to the nonexistence of
other renewable resource data in the selected location. For example, this concerns the flow
rate data of wind streams and the amount of biofuels available throughout the year.
This study will use HOMER software for modelling and simulation.
1.7 Analysis Framework
The concept of ‘analysis of power system options for rural electrification’ is increasingly important
for the developing countries. The figure 1.1 shows the framework for analysing the hybrid system
or combination of RETs for electrification of rural villages in Rwanda.
The framework shows how, in different contexts, the best techno-economic combination of RE
resources are achieved through the modelling and simulation using HOMER software to combine
the input data; the load profile, renewable energy resources and the equipment’s cost for best
configuration.
The key components of this project is shown in the framework as the analysis of the initial site
assessment, details assessment, data bank analysis, system design, techno-economic analysis and
end up with the best techno-economic combination of RE resources in a hybrid power system for
electrification to the selected community in Rwanda.
6
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
INITIAL ASSESSMENT
-Existing situation
-User needs & Demand
-Energy Resources
Budget & Finance Availability
GRID EXTENSION
INITIAL LOADING DATA
COOKING
Wood/LPG
ELECTRIC
SELECTION OF ENERGY
SOURCES
NON-ELECTRIC
RENEWABLE ENERGY
OFF-GRID ELECTRICITY
HEATING
Solar/Wood/LPG
DETAIL ASSESSMENT
LOADS
-User profile
-Daily Load
-Priority loads
Site Characteristics
-No. of houses
-Population
-Area details
DATA BANK-ANALYSIS
SYSTEM DESIGN WITH HOMER
SOFTWARE
-Configuration
-Generation method
Too expensive
TECHNO-ECONOMIC ANALYSIS OF
THE SYSTEM WITH HOMER
Cost Competitive
SELECTING BEST DESIGN SYSTEM
FROM HOMER SIMULATION
Approved
DETAILED FINANCIAL ANALYSIS
Figure 1.1 : Framework of analysis
7
RESOURCES
-Solar/Small-Hydro
-Geographical & Meteorological
data
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
1.8 Thesis Outline
Chapter 2 reviews the load profile and available resources in the village location, hydro resource,
solar resource and the climate data of the village. Chapter 3 will be concerned with the explanation
of the major components used in renewable energy technology system. It illustrates the important
characteristics of the system components such as electrical characteristics, costs, operation and
maintenance difficulties. Chapter 4 discusses the modeling of the hybrid system in HOMER
software. Chapter 5 discusses the results obtained from the simulations of the hybrid system in
HOMER software. The results of the optimization and sensitivity analysis, the selection of the
optimal hybrid configuration and the performance of the selected system for varying conditions of
load, solar and hydro resource will be discussed in this chapter. Chapter 6 then presents the
discussion, while concluding remarks and future work are presented in Chapter 7.
8
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
2 Data Collection
Hybrid system design and optimization requires an evaluation of the load profile of the village and
the renewable resources in the region. In this chapter we are going to discuss the estimation of
village load profile and the assessment of renewable resources, solar and hydro at the site. The
chapter discusses calculation of solar radiation on a tilted PV panel using horizontal radiation data
and the monthly average water flow will be carefully estimated based on the average precipitation,
average temperatures and topography of the region.
2.1 Introduction
One of the villages from the Burera District in North Province, Rwanda is selected for analysis of
option of renewable hybrid energy system for supplying electricity. The map of the Burera district
is given in Figure 2.1. Burera district consists with area of 644.5 km² and density of 522.2 inh./km².
The EICV3 survey results show that the total population of Burera district in 2010–2011 was
354,000. This represents 18 % of the total population of Northern Province and 3.3 % of the total
population of Rwanda [5].
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. In Burera
district, only 3.2 % of households use electricity as their main source of lighting, ranking the district
third ranked after Musanze (14.5 %), Gicumbi (8.9 %) in Northern Province. The urban area average
is 46.1 % of households using electricity as their main source of lighting, while it is only 4.8 % in
rural areas and 10.8 % at national level. Hence Burera district is below the national, urban and rural
area averages [5].
Figure 2.1: Map of Burera District [13].
9
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
In this research, a village located at 1o30’ S latitude and 29o58’ E longitude has been selected for
placement of the hybrid system. The geography of the selected village is presented in Figure 2.2
and 2.3. As presented in Figure 2.3, the electrical loads are scattered all over the village.
Figure 2.2 : Map of Geography allocation of Karegamazi site [14].
Figure 2.3 : Closer or zoomed view of Karegamazi village [14].
10
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
In order to assess the applicability of a hybrid RES for supplying electricity, firstly it is required to
discover the potential of RE resources in the selected area and the demand for the electricity of the
selected community [15].
2.2 Village Load Profile
In a remote rural village the need for electricity is not high as match up to urban areas. Electricity
requirement is for domestic use (for appliances such as radio, color television, compact fluorescent
lamps, DVD player, refrigerator, computer, and an iron, community activities (such as in
community halls, schools and health post) and for rural commercial and small scale industrial
activities (such as cold storage, small processing plants for cassava flour and sorghum flour and
cottage industries).
A survey in the village will be required to conduct for collecting all these data. But real surveyed
data is not available for the selected community, the load profile of the village has been derived
based on the knowledge that I have on the selected area and assumptions by using the results
obtained from the interviews with the households which have been conducted on the new
community area where the power extension have been reached. Survey form for Households can be
found in appendix A.
The selected village consists of 10 rich families, 40 medium income families, 100 low income
families and 50 very poor families, the latter being excluded in this regards. The village has 5 shops
and bars, two administration posts, one medical center, one primary and one secondary schools, one
community church and 3 small manufacturing units. The detailed daily consumption for selected
village and the daily power hourly distribution can be seen in the appendix B and C respectively.
To be more specific concerning “rich”, “medium”, “low income” families; according to Andrew
Kettlewell, the Adviser of Technical Team for Rwanda’s Vision 2020 Umurenge Programme also
known as VUP;
Rich families are those which have land and livestock, and usually have jobs where they can earning
some money. Good housing, generally own a motorbike or vehicle, and people who can do business
with bank so that they can easily get credit from the bank [16].
Medium income families are those with larger landholdings on productive soil and sufficient to eat.
Own livestock, sometime they have a small paid jobs, and can have access to health care [16].
Low income families are those which have very small land and small house. Live on their own labor
and even if they don’t have some savings, they can find something to eat, even though the food is
not very healthful and some of them their children may go to primary education [16].
Very poor families are those which have to beg for surviving, no land, no livestock and no safe
house and no adequate dress and food. They don’t have access to medical care due to the lack of
money and the government have to pay for them. Their Children do not attend school. But some of
them may be physically capable to work in the land owned by others and earn some money for
nourishment [16].
11
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Table 2.1 : Assumptions on daily consumption for the selected community.
No
1
2
3
4
5
6
Consumers type
Rich families
Medium income families
Low income families
Shops and bars
Administration posts
Medical center
Number
10
40
100
5
2
1
Daily consumption in kWh
46
32
39
35
3
34
7
Primary school
1
5
8
Secondary school
1
11
9
10
Community church
Small manufacturing units
1
3
5
49
Based on these, a typical daily load curve with hourly resolution has been derived for this village
and it is given in Figure 2.4. With respect to the derived load profile, the maximum demand of the
village is around 28 kW but with the random variability of 10 % (standard deviation: daily and
hourly noise to make the load data more realistic) for both day to day and time step to time step,
this maximum demand can become 38 kW with the energy consumption of around 249 kWh.
30
Power (kW)
25
20
15
10
5
00:00-01:00
01:00-02:00
02:00-03:00
03:00-04:00
04:00-05:00
05:00-06:00
06:00-07:00
07:00-08:00
08:00-09:00
09:00-10:00
10:00-11:00
11:00-12:00
12:00-13:00
13:00-14:00
14:00-15:00
15:00-16:00
16:00-17:00
17:00-18:00
18:00-19:00
19:00-20:00
20:00-21:00
21:00-22:00
22:00-23:00
23:00-24:00
0
Day-hours
Figure 2.4 : Village load profile
2.3 Solar Resource Assessment
For assessing the option of using solar (photovoltaic) power, we have to consider the solar resources
in our simulation. The resource assessment is presented below. As there is a long distance from the
selected village to the next weather station where ground measurements of solar radiation are
performed, the solar resource information used for selected village at a location at 1o30’ S latitude
12
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
and 29o58’ E longitude was taken from the NASA Surface Meteorology [12] as made available by
RET Screen International [12]. Data on the monthly averages of the daily radiation sum on a
horizontal surface are plotted in Figure 2.5. In addition, tabulated monthly averaged daily insolation
incident are given in Table 2.2 together with the clearness index [17]. The clearness [4] is a measure
of the fraction of the solar radiation that is transmitted through the atmosphere to the earth's surface.
The annual average solar radiation was found to be 5.13 kWh/m2/day and the average clearness
index was found to be 0.513.
When comparing the selected village in Rwanda and a village in central Germany, the village
Niederdorla, located at 51°09’ N latitude and 10°26’ E longitude, as show on Table 2.3 from the
NASA website [12], it shows that the annual average solar radiation of Niederdorla village is 2.72
kWh/m2/day and its average clearness index is 0.39.
By comparing both results shown in both Tables 2.2 & 2.3, it is clear that, the selected village in
Rwanda has both quiet good solar radiation and clearness index than Niederdorla village in
Germany. Due to that solar radiation data, it is clear that the average solar radiation in Burera village
is relatively good. This would give an approximately good probability and occasion to use the
photovoltaic technology as one element of the hybrid RES.
The monthly mean temperatures of the village located at 1°30’ S latitude and 29°58’ E longitude is
given in Table 2.4. They range from 21.6 °C to 24.5 °C throughout the year. Thus this area is not
affected by seasonal variations. Also the day length in Rwanda does not vary throughout the year
due to its geographical location. Due to the small variations of irradiance and temperature, there it
is expected that there are no significant changes in the load curve within the year. Therefore, a
constant daily load profile has been assumed for the entire year.
Figure 2.5 : Monthly radiation sums for the selected village, from Homer.
13
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Table 2.2 : Monthly average daily irradiance incident on a horizontal surface for the target
location [12].
Month
January
February
March
April
May
June
July
August
September
October
November
December
Average
Clearness Index
0.557
0.569
0.525
0.515
0.542
0.516
0.49
0.483
0.507
0.464
0.477
0.51
0.513
Daily Radiation (kWh/m2/d)
5.69
5.97
5.52
5.22
5.16
4.72
4.55
4.74
5.23
4.84
4.88
5.139
5.133
Table 2.3 : Monthly average daily irradiance on a horizontal surface for Germany [12].
Month
Daily Radiation (kWh/m2/d)
Clearness Index
Jan
0.36
0.84
Feb
0.39
1.54
Mar
0.39
2.42
Apr
0.41
3.64
May
0.42
4.58
Jun
0.41
4.78
Jul
0.42
4.66
Aug
0.44
4.15
Sep
0.39
2.78
Oct
0.35
1.64
Nov
0.32
0.89
Dec
0.34
0.65
Average
0.39
2.72
14
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Table 2.4 : Monthly mean values for other climatic parameters in Burera District [12].
Month
January
February
March
April
May
June
July
August
September
October
November
December
Annual
Air temperature
°C
23.8
24.5
23.4
22.5
22.4
22.6
23.1
22.4
21.9
21.6
21.7
22.5
22.7
Relative
humidity %
53.4
52.9
68.3
77.0
74.0
65.9
56.7
66.3
74.8
79.0
77.2
65.2
67.5
Atmospheric
pressure kPa
89.8
89.8
89.7
89.8
89.9
90.0
90.0
90.0
90.0
89.9
89.9
89.8
89.9
Earth
temperature °C
24.1
24.9
23.8
22.7
22.4
22.4
23.0
22.5
22.0
21.9
21.7
22.4
22.8
2.4 Hydro Resource Assessment
D. Magoma, P.M. Ndomba, F. W. Mtalo, and J. Nobert [18] in their research for Rugezi catchment
situated in the Northern province of Rwanda in Burera district, have shown that the rugezi
catchment has about 196 km² (Figure 2.6). It is located between latitudes 1o21’30” and 1o36’11”
South and longitudes 29o49’59” and 29o59’50” East. The Rugezi catchment divided into sub
catchments: The Rugezi main (164 km²) and the Kamiranzovu watershed (32 km²). The main
Rugezi is situated in the east of Lakes Burera and Ruhondo below the Virunga volcanoes. They
have done two test and compere them; simulated test and observed stream flows test at the drainage
area, Rusumo gauging station [18], the results in Figure 2.8 where the data for calibration period
1976-1981 (four years) shows that; the observed day to day flow for this period of 4 years is 1.38
m3/s and the simulated is 1.31 m3/s [18].
15
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 2.6 : Placement of Rugezi catchment in Burera District [18].
Figure 2.7 : Reservoir of karegamazi at which the hydropower plant is possible [14]
16
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 2.8 : Discovered and simulated daily stream flow [18].
Since the above data is not sufficient for the assessment because the Homer software will require
the monthly water stream flow in liter per second , I have tried to search for other information so
that I can compare the results from [18] Figure 2.8 and the current hydro resources for Rugezi. The
current information of water stream flow at Rusumo gauging station is shown on the Figure 2.9,
source from the Rwanda Energy Group (REG) by E-mail correspondence.
These data have been obtained by the recording from Rugezi Micro Hydro Power Plant which was
working in the day of 2012, unfortunately this plant has stopped due to the wrong design and
construction.
4000
3500
Litre per Sec.
3000
2500
2000
1500
1000
500
0
1
2
3
4
5
6
7
8
9
10
11
12
Months
Figure 2.9 : Average monthly stream flow at Rusumo gauging station.
17
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
With the data above show that the annual flow is maximum in April with a stream flow of 3390
litres per second and the minimum is in the month of august with a stream flow of 1150 litres per
second. The annual average water release from Rugezi catchment is 2019 litres per second, and the
residual flow which is around 350 litres per second is not included. The residual flow is part of
water which is undistributed in the plant for ecological causes to support fish populations [4].
In this project, it is assumed that only a small portion of this water can be used. As given in figure
2.7 of the reservoir where the micro hydropower plant is possible, is situated in the east of the
Rugezi main catchment. Since water from the Rugezi catchment is the source of two other lakes;
Burera and Ruhondo and those lakes are the source of other big two hydro power plants which are
Mukungwa and Ntaruka respectively. As explained in the following, this situation sets limits to new
hydro power stations.
There are some rules and regulations from the ministry of environment, lands and mines required
to use all kind of activity related with the Rugezi catchment. This is due to vulnerability which has
taken cover in the year of 2000, when the country has passed through the crisis of electricity supply
and due to this Rwanda has been negatively affected in many aspects[19]. The trouble stimulated
by an extreme reduction of power generated from Ntaruka power station and that one from
Mukungwa power station, and at that time, these two power stations cover 90 % for the whole
country power demand. The decreases of electricity generation from Ntaruka and Mukungwa has
been affected by the drop of water in Lake Burera the reservoir of the two stations [19]. This water
drop has been affected by several factors, like; bad management for surrounding of the Rugezi
watershed caused by human activity and technical problem connected with bad maintenance of
stations.
Due to the above reasons, even if there is a water flow of 2019 litre per second from the rugezi
catchment, in this project 280 litres per second will be used with the head of 9.7 metres to produce
the output power approximated to 20kW.
As the typical forecasted load profile of the selected community and the identification of the
possible renewable energy resources was presented, it is good idea to see and discuss the system
layout caused by the fact that the maximum load is more than 20 kW and Off-grid renewable energybased power systems cannot provide a continuous supply of electricity without a storage medium,
consequently batteries are added to the hybrid system. In order to ensure the continuity of the
supply, a diesel generator are also incorporated. Further, various component configurations for the
system have to be characterised.
18
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3 Hybrid System Components Characteristics and Costs
In this chapter we will discuss the characteristics, operation, maintenance and the relevant costs of
the hybrid system components. We will start by discussing the basic technological configurations
of hybrid systems. Then the chapter explains the characteristics of the components; PV panel, Micro
hydropower system, diesel generator, storing bank and the inverters. These are the relevant
components used in the hybrid system studied in this project.
3.1 Introduction
In this HOMER analysis, solar PV, and run-off river micro hydro power are the principal resources
and the diesel is used for the emergence cases. Batteries and converter will be used for storing and
converting from one form to other form system of electricity, respectively. The performance and
cost of each of the system’s components is a major factor for the cost results and the design.
Depending with the kind of voltage system and bus that interconnect the sources, there are many
different types of hybrid system,



DC coupled system,
DC/AC coupled system
AC coupled hybrid system
In this study, I prefer to use the AC coupled hybrid system where all electricity generating sources
are connected to the AC bus because of the following reasons:
DC coupled Hybrid system all sources are networked to the DC bus. This means that the PV
generating source is equipped with charging controller and AC generating sources with rectifiers,
this means that the power generated by the diesel generator and the alternator are first rectified and
then converted back to AC which reduces the efficiency of energy conversion due to several power
processing stages. Due to this reason, the DC coupled hybrid system have not been selected for this
study.
In DC/AC coupled hybrid system, electricity generating sources can be connected to either DC or
AC bus depending with the generating voltage form. This hybrid system uses a bidirectional inverter
to link the DC bus and the AC bus. Also the efficiency of the generator can be maximized due to
the capability to operate the inverter in parallel with the AC sources. Unfortunately this system have
not been selected due to its two buses and to ignore the danger which may be generated due to
failure of the bidirectional inverter.
In AC coupled hybrid system the DC generating sources are linked to the AC bus through inverters
and AC sources can be immediately bridged to the AC bus or maybe through a medium to facilitate
stable link. Regarding the battery bank, the energy supply is controlled by a bidirectional inverter.
AC coupled systems is more flexible, easily expandable and it offer a flexibility for grid extension
when necessary [4]. Due to the above functionality, this type of system has been selected for this
project.
19
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Since the AC coupled hybrid system has been selected, as it is shown on the Figure 3.1 the main
components for the system are the follows; PV panels, Micro hydro power plant, batteries, diesel
generator and inverters. In this project, two inverters have been used for a solar inverter and a
bidirectional battery inverter. That is why this chapter will discuss each of this component’s
functionalities, specifications and costs.
AC BUS
Hydro Power Plant
Diesel Generator
Solar Inverter
PV Array
AC Load Home Houses
Bi-directional Inverter
Battery Bank
Figure 3.1 : AC coupled hybrid system.
3.2 PV Panels
Solar system is the greatest and favorable of the renewable sources because of its apparent indefinite
potential [1]. The sun emits its energy and the latter is transmitted as electromagnetic radiation, the
letter can be used by photovoltaic module to produce a direct current. After the sun radiation being
passed through the atmosphere, 1kW of solar power can be experienced on an area of one square
meter [20]. The output power from a typical solar cell is around 1 watt. That is why to generate the
required amount of power a certain number of cells are connected in compound in order to have a
complete module.
20
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3.2.1 Electrical characteristics of PV cells
A perfect solar cell is presented by the combination of a current source connected in shunt with a
diode[21]. Its equivalent I-V characteristic is calculated by the equation (3.1) [21][22].
𝐼 = 𝐼𝑝ℎ − 𝐼𝑜 (𝑒
𝑞𝑉
𝑘𝐵 𝑇
− 1)
(3.1)
Where
kB : Constant of Boltzmann,
T : Absolute temperature,
q (>0) being electron charge,
V the voltage of the cell and
Io is the diode saturation current.
A solar cell act as a diode during the darkness. Figure 3.2(Top) shows the I-V characteristic of
Equation (3.1). In theoretical, the Isc is equal to the photo generated current Iph, and open voltage
Voc is given by
𝑉𝑂𝐶 =
𝑘𝐵 𝑇
𝑞
𝑙𝑛(1 +
𝐼𝑝ℎ
𝐼0
)
(3.2)
The power produced by the solar cell is shown in Figure 3.2(Bottom) [21]. The cell generates the
maximum power Pmax and it is appropriate to calculate the fill factor FF by
𝐼 𝑉
𝑃
𝐹𝐹 = 𝐼𝑚𝑉𝑚 = 𝐼 𝑚𝑎𝑥
𝑉
𝑠𝑐 𝑜𝑐
(3.3)
𝑠𝑐 𝑜𝑐
The Figure 3.2 below shows the I-V characteristic of an perfect solar cell (Figure 3.2 top) and the
power produced (Figure 3.2 bottom) and the power at the maximum power point is the shaded
rectangle in Figure 3.2 top [6].
Figure 3.2 : The I-V and Power aspect of a perfect solar cell [22].
21
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
The I-V nature of a solar cell in practice normally has some difference with the ideal characteristic
[21][22]. A two-diode model is often used to be able to obtain an observed curve, with the second
diode has an ideality factor of two in the denominator of the argument of the exponential term [21].
Its circuit may also have series (Rs) and parallel (Rp) resistances, conduction to the following
equation [21].
𝐼 = 𝐼𝑝ℎ − 𝐼01 {𝑒
𝑉+𝐼𝑅𝑠
𝑘𝐵 𝑇
𝑉+𝐼𝑅𝑠
− 1} − 𝐼02 {𝑒 2𝑘𝐵 𝑇 − 1} −
𝑉+𝐼𝑅𝑠
𝑅𝑝
(3.4)
where the light-generated current Iph may, in some cases, depend on the voltage [21]. These
features are presented in the equivalent circuit in Figure 3.3 by the dotted lines [21]. The effect of
both resistance and the second diode on the I-V characteristic of the solar cell is presented in Figures
3.4 and 3.5, respectively [21]; further information see the Figure 3.6.
Figure 3.3 : The equivalent circuit of non-ideal solar with components in dotted line [22].
Figure 3.4 : The I-V characteristic of PV in the two diode model [22].
Figure 3.5 : The effect of resistance on the I-V characteristic of PV [22].
22
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
.
Figure 3.6 : The dark I-V characteristic of PV in the two diode and series resistance [22].
The power produced by a crystalline PV module is affected by two key parameters;
 Solar irradiance
 Cell temperature
The effect of the solar irradiance and the module temperature on the I – V characteristic of the
German Solar GSM6-250P, the information from the datasheet as presented in Figure 3.7 shows
that the output current of the cell drops when the solar irradiance level decreases. The same case
take cover for the output power which decreases also but the open circuit voltage is not much
affected. In case of temperature this happen in opposite where open circuit voltage decreases with
the increases of temperature in the module but this does not affect significantly on the short circuit
current. The German Solar GSM6-250P have been used for the explanation but this happen for all
the kind of solar cells.
Figure 3.7 : Effect of solar irradiance and cell temperature on the I–V curve [23].
23
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3.2.2 Operating Temperature of PV cells
Solar irradiance on the solar cell is the cause of its electrical power output put also causes a heating
up of the module. For the good working condition, the cells should work on the minimum possible
temperature. An energy balance on a unit area of module can be used to find out the temperature at
which the cell should operate [6]. This is obtained by the equation 3.5 [6][24].
𝜏𝛼𝐺𝑇 = 𝜂𝑐 𝐺𝑇 + 𝑈𝐿 (𝑇𝑐 − 𝑇𝑎 )
(3.5)
Where
𝜏: The solar transmittance of the cover in percentage
𝛼: The solar absorptance in percentage
𝐺𝑇 : The solar radiation striking the array (kW/m2)
𝜂𝑐 : The electrical efficiency of array in percentage
𝑈𝐿 : Heat transfer coefficient (kW/m2 0C)[4][6]
𝑇𝑐 : The temperature of the cell (0C)[4][6]
𝑇𝑎 The ambient temperature (0C)[4][6]
To characterize the heating up of the module due to irradiance. The cell temperature for steady state
conditions under constant irradiance and temperature can be measured. According to US-standards,
the cell temperature should be measured at 800 W/m2 and an ambient temperature of 20ºC called
the nominal operation conditions NOCT.
Measurement of cell & ambient temperature, and solar radiation can be used for calculating the
ratio 𝜏𝛼/𝑈𝐿 [24]
𝜏𝛼/𝑈𝐿 =
𝑇𝑐,𝑁𝑂𝐶𝑇 −𝑇𝑎
(3.6)
𝐺𝑇,𝑁𝑂𝐶𝑇
Where
𝑇𝑐,𝑁𝑂𝐶𝑇 : The Nominal Operating Cell Temperature (0C)[4][6]
𝑇𝑎 : The ambient temperature for NOTC is defined (20 0C)[4][6]
𝐺𝑇,𝑁𝑂𝐶𝑇 : The radiation of solar with NOCT is defined (0.8 kW/m2)[4][6], this is for
standard of USA characterization for solar module. Homer also use this as input variable.
By considering the ratio 𝜏𝛼/𝑈𝐿 to be constant, the temperature at any other condition can be
calculated with
𝜏𝛼
𝜂𝑐
𝐿
𝜏𝛼
𝑇𝑐 = 𝑇𝑎 + 𝐺𝑇 ( 𝑈 )(1 −
)
(3.7)
The 𝜏𝛼 is not known in most of the case but this can be approximated to be 0.9 because the
ratio 𝜂𝑐 /𝜏𝛼 is so small than a unity.
24
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
When solar operate on its MPP the PV efficiency is the efficiency at MPP [4].
𝜂𝑐 = 𝜂𝑚𝑝𝑝
(3.8)
Since the efficiency at MPP changes with the changes of the cell temperature then the variation can
be calculated as follows
𝜂𝑚𝑝𝑝 = 𝜂𝑚𝑝𝑝,𝑆𝑇𝐶 {1 + 𝛼𝑃 (𝑇𝑐 − 𝑇𝑐,𝑆𝑇𝐶 )}
(3.9)
Where
𝜂𝑚𝑝𝑝,𝑆𝑇𝐶 : The MPP efficiency under the test at standardized conditions (%)
𝛼𝑃 : The temperature coefficient (%/0C)[4][6]
𝑇𝑐,𝑆𝑇𝐶 : The cell temperature under the test at standardized conditions (250C)
Using equations, 3.6, 3.8 and 3.9 and put into equation 3.7, the temperature of the cell at any
irradiance can be obtained with the equation (3.10)[6].
𝑇𝑐 = 𝑇𝑎 + 𝐺𝑇 (
𝑇𝑐 =
In practice, as the
𝜏𝛼
𝑈𝐿
𝑇𝑐,𝑁𝑂𝐶𝑇 −𝑇𝑎
𝐺𝑇,𝑁𝑂𝐶𝑇
)(1 −
𝜂𝑚𝑝𝑝,𝑆𝑇𝐶 {1+𝛼𝑃 (𝑇𝑐 −𝑇𝑐,𝑆𝑇𝐶 )}
𝜏𝛼
𝜂𝑚𝑝𝑝,𝑆𝑇𝐶(1−𝛼 𝑇
𝐺𝑇
𝑃 𝑐,𝑆𝑇𝐶 )
){1−
}
𝐺𝑇,𝑁𝑂𝐶𝑇
𝜏𝛼
𝛼
𝜂
𝐺𝑇
𝑃 𝑚𝑝𝑝,𝑆𝑇𝐶
1+(𝑇𝑐,𝑁𝑂𝐶𝑇 −𝑇𝑎,𝑁𝑂𝐶𝑇 )(
)(
)
𝐺𝑇,𝑁𝑂𝐶𝑇
𝜏𝛼
)
(3.10)
𝑇𝑎 +(𝑇𝑐,𝑁𝑂𝐶𝑇 −𝑇𝑎 )(
(3.11)
in this formula are not known from standard module test 3.11 is replaced by
𝑇𝑐 = 𝑇𝑎 + 𝑐 ∗ 𝐺 [25]
(3.12)
With c being a constant reflecting the type of module mounting (freestanding, roof integrated,…),
see e.g in [25].
3.2.3 PV module Power output
The power output of a PV as it has been discussed that it is a function of the temperature and the
irradiance of the solar and can be found by equation 3.13 where cell temperature is calculated as it
has been proved in the equation 3.7.
𝑃𝑃𝑉 = 𝑌𝑃𝑉 𝑓𝑃𝑉 ( 𝐺
𝐺𝑇
𝑇,𝑆𝑇𝐶
)(1 + 𝛼𝑃 (𝑇𝑐 − 𝑇𝑐,𝑆𝑇𝐶 ))
(3.13)
Where
𝑌𝑃𝑉 : is the module rated capacity (kW)
𝑓𝑃𝑉 : is [6]the module derating factor (%), HOMER exercises this factor to the output power
PV
array to take into account some factors which lower the output in real
25
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
conditions [4][6]. Such factors may be
snow cover, shading, and so on.
dusty of the panels, network losses,
aging,
𝐺𝑇,𝑆𝑇𝐶 : is the incident radiation under the test at standardized conditions (1 kW/m2) [4]
𝑇𝑐 : is the cell temperature (0C) [4]
3.2.4 PV cost
Photovoltaic Solar panels cost has been reduced drastically in the past years and it is assumed to
continue its down slope for the future; the cost of solar panels is a variable that actually depends on
the time, place and scale of the solar panel installation.
According to the reported pricing for PV system installations, the current overall cost figures in
recently updated prices are as follows [26]:
• Residential and small commercial (≤10 kW) was $ 4.69 /W (median)
• Large commercial (>100 kW) was $ 3.89/W (median)
• Utility-scale (≥5 MW, ground-mounted) was $ 3.00/W (capacity weighted average).
PV modules certified for conformity with the IEC61215 (Crystalline silicon terrestrial photovoltaic
(PV) modules – Design qualification and type approval) standard for the mono-crystal and with
similar IEC standard for the poly-crystal, the costs are given for a 10 kW fixed slope PV system.


The price of Monocrystalline Solar Panel SUNTECH STP250S is 245.63 € [27] equal to US
$ 360. The 10kW will cost $ 360*40 = $ 14400, considering transport of 20% and taxes of
18%, the total cost for 10 kW comes to $ 20000
The cost of solar inverter is $0.435/Wp [28] this means that the cost of 10kW will be $ 4350,
by considering transport of 20% and taxes of 18%, the total for 10 kW will be $6000.
Balance of System Cost
 The estimated cost for the solar ground mounting system is $ 100 per module, since the
module of 250 Wp have been selected, then the cost for the 10 kW which is 40 modules of
250Wp system is $ 4000.
 The Local transportation cost of the equipment from Kigali to Burera is estimated as $ 500.
 The estimated installation cost and other relevant cost is $4500
The total costs is around $ 35000 which is the estimation costs for 10 kW solar PV system. Solar
system do not require a lot of maintenance work as compared with other technologies with moving
parts. Thus the operating and maintenance cost of a PV system is relatively small. The annual O&M
cost of a 10 kW PV system has been considered as $ 30.
26
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 3.8 : Solar PV ground mounted system [29].
3.3 Micro-Hydro Power Plant
It is a non-polluting and environmental friendly source of energy. Hydropower is established with
simple concepts. Water movement rotates a turbine which is mechanically connected to generator,
and electricity is produced. Many other components are required, but it all starts with the energy
from water. The use of water falling through a height has been utilized as a source of energy a very
long time [30].
3.3.1 Components Overview
Figure 3.9 presents the principal elements of a run-of-the-river micro-hydropower system. As the
Figure shows, no storage of water but instead the pipe connect the river and the penstock, then the
latter connect the stream of water to the turbine. The power poles or tower can be used to transmit
the power from the power plant up to end users [31][30].
27
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 3.9 : Micro hydropower plant overview [30].
Many aspect can be used to build up a micro hydro power plant depending in accordance with the
geographic and hydrological conditions, but general concept is the same.
The following figures are the principal components of a run-of-the-river micro-hydro system [32].

Diversion Weir and Intake
The diversion weir is a block barrier constructed over the river and it is used to redirect the
water through the ‘Intake’ opening into a settling basin.
Figure 3.10 : Diversion Weir and Intake [32]

Settling Basin
The settling basin help to filter the water before entering the penstock. This can be
constructed at the intake or at the forebay.
28
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 3.11 : Settling Basin [32]

Headrace
A conduct that govern the water to a forebay or turbine. The headrace pursue the contour
of the hillside so as to maintain the elevation of the diverted water.
Figure 3.12 : Headrace [32].

Head tank
Small reservoir at entrace of a pipeline; this is taken as final settling basin, provides
overflow of penstock inlet and integration of trash rack and overflow/spillway
arrangement.
29
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 3.13 : Head Tank [32].

Penstock
An enclosed conduit which is used for furnish the pressurized water to a hydro turbine.
Figure 3.14 : The penstock [32].

Water Turbine and alternator
A turbine is a machine converting the kinetic energy of water into a rotational energy at
the same time, the alternator is another electrical machine for converting mechanical
energy into electrical energy.
Figure 3.15 : Connection arrangement between Turbine and Generator [32].
30
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3.3.2
Micro Hydropower Capacity
For information about the power potential of water in a stream, it is very important to know the
quantity of water flow available from the stream (for power generation) and the available head. The
available water for power generation is the amount of water (in m3 or litres) which can pass via an
intake into the pipeline (penstock) in a given amount of time. This is normally expressed as (m3/s)
or in litres per second (l/s). The head is the vertical difference in level (in meters) through which
the water falls down. The theoretical power (P) can be calculated using the following equation
[31][32].
P= Q × H × e × 9.81 Kilowatts (kW)
(3.14)
Where
P: Generator Output Power (kW)
H: The water head in metres (m)
Q: The water flow (m3/s)
e: The total efficiency (%)
g: 9.81 is a constant
The output power will be the function of several loss which will take cover in the production system
as indicated in the figure 3.16.
Figure 3.16 : Typical system losses for a system running at full design flow [32].
3.3.3 Micro Hydro Power Plant Cost
While performing a trial calculation of construction cost in the planning stage, this can be done by
following some method. However, before the calculation, it is necessary to carry out a field survey
for confirmation and decide the item mentioned in the table 3.1 below [32].
31
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Table 3.1 : Items to make a trial calculate of construction cost.
Description
Plan
Intake Facilities
Headrace
Penstock
Distribution
Item
Maximum Out Put (kW)
Turbine Discharge (m3/s)
Effective Head (m)
Height of Dam (m)
Length of Dam (m)
Length of Headrace (m)
Diameter of Penstock (m)
Number of Households
Distance to the most far house from P.S
In addition to the direct costs, indirect costs, such as Tax, Contractor fee, Design Cost, and
Supervision cost, are contained in the cost of construction. When part of these indirect costs is
missing, some explanation is required separately [32].
Items of direct cost
Typical items of a direct cost are the following [32].
1. Preparatory Works
Preparatory Works consist of item as follows.
 Location Setting Out,
 Filling and Measurement,
2. Civil Works
Civil Works consist of item as follows
 Intake facilities,
 Settling basin,
 Headrace,
 Head tank,
 Spillway,

Equipment & Materials
Mobilization




Penstock and Foundation,
Powerhouse base,
Tailrace,
Power house,
3. Electro-Mechanical Works
Electro-Mechanical Works consist of item as follows.
 Turbine,

 Controller,

 Dummy load,

 Generator,
4. Distribution Works
Distribution Works consist of item as follows.
 Transmission pole,
 Distribution Wires,
32


Accessories,
Spare parts and Tools
Set up and Installation
Step up/down Transformer,
Other extra
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Quantities
In order to know the direct cost of construction for MHPP, it is required to know the quantity for
every work or material based on the design. For example, in case of Headrace made of stone
masonry, quantities of excavation, foundation rubble stone, stone masonry, backfill, and plastering
shall be estimated.
Unit Cost
Since the cost of micro hydro power plant differs according to various items in which it is very hard
to know the cost of every one because most of the items require a lot of understanding, in this project
I prefer to estimate the cost of micro hydro power plant using other researches which have been
demonstrated for the cost per watt of the output power produced.
In Renewable Energy – Based Mini – Grid for Rural Electrification: Case Study of an Indian
Village[9], Rohit Sen and Subhes C. Bhattacharyya in their research, they have demonstrated that
the capital cost for a 30-kW SHP can be assume as $42,000 while the replacement cost and O&M
cost are considered to be $35,000 and $4,000, respectively. I will use the same approach in my
project because, this is true when using the information from [33] saying that, internationally an
initial capital cost estimated for micro hydro power plants, with new technologies, is estimated in
between US$ 1500 to $ 2500/kW where this cost is composed with around 75% of the development
cost and it is decided by the location conditions, and the remaining 25% is the cost of purchasing
engineering components(the turbine, generator, electronic load control, manual shunt-off valve, and
other components) [15].
In “Economic Analysis and Application of Small Micro- Hydro Power Plants” by Mrs. Sarala P.
Adhau[34] state that, the investment cost for a micro hydro power plant can be estimated as $ 1500
per kW.
In this development, the cost is taken as an average at $ 1500/kW because of the remote area, and
thus complicated position of the village and neighbouring areas [15].
A design flow rate of 280 l/s at 9.7 metres head, a turbine coupled to an alternator will be able to
produce an electrical output power of 20kW, at an overall efficiency of 75%.
The capital, renewal, and O&M worth of the micro-hydropower system were estimated at $ 40000,
$ 30000 and $ 800 /year respectively.
3.4 Diesel Generator
Diesel generators are very important in renewable energy hybrid systems to improve the quality and
the availability of the electricity supply. Since diesel generators is used for extra load in an occasion
of necessity, or in a case of black-out and battery bank is not sufficient for supplying the load [17].
The initial investment for a diesel generator is relatively small when compared with the initial
investment of the micro hydro power plant, PV or wind power plant system. The big problem with
this is that, the operating and maintenance cost of a diesel generator is very high because it
33
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
requires a continuous supply of fuel (diesel), and frequent maintenance and inspection of the
engine throughout its operating life.
Normally diesel generators operate most efficiently when working near its full load. That is why it
is good to work the generator beyond a certain load factor for controlling the proper efficiency of
the energy conversion and then lowering the fuel cost by reducing the utilization of fuels. The fuel
consumption curve of a diesel generator is normally given in the product data sheet and the
efficiency curve can be derived using that product specification.
The Figure 3.17 below represents a typical genset fuel curve. Diesel generators are most efficient
when used near its highest output. As well as the load decreases, the efficiency will decrease
significantly. When increasing the load from 20% to 80% doubles the efficiency of the generator,
reducing fuel consumption per kWh by two. The diesel fuel consumption is given in a Table 3.2
depending with the power of the generator and the percentage of its full load.
Figure 3.17 : Typical generator efficiency curve [35].
Table 3.2 : Approximate Diesel Fuel Consumption Chart [36].
Generator Size
(kW)
20
30
40
60
75
100
125
135
150
175
200
1/4 Load
(litre/hr)
2
5
6
7
9
10
12
12
14
16
18
1/2 Load
(litre/hr)
3
7
9
11
13
16
19
20
22
26
29
34
3/4 Load
(litre/hr)
5
9
12
14
17
22
27
29
32
37
42
Full Load
(litre/hr)
6
11
15
18
23
28
34
37
41
48
55
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3.4.1 Operation and Maintenance of Diesel Generators
The lifetime of a diesel generator is operation hours and it depends on various aspects. When the
generator will be used in standby power applications, it is assumed to start at full-rated load in less
than 10 seconds. However the maintenance of generator like other mechanical device is a paramount
for ensuring that a diesel powered standby generator will start and run when desired [37].
To manage the efficient operation and expand the operating life of diesel generators it is required
to provide and obey to the maintenance schedule based on the specific power application and the
severity of the environment [37]. The Table 3.3 below gives the required regular and typical diesel
maintenance schedule and their estimated costs:
Table 3.3 : Regular and typical diesel maintenance schedule and their estimated costs.
Maintenance Items
Inspection
Check coolant heater
Check coolant level
Check oil level
Check fuel level
Check charge-air piping
Check/clean air cleaner
Check battery charger
Drain fuel filter
Drain water from fuel tank
Check coolant concentration
Check drive belt tension
Drain exhaust condensate
Check starting batteries
Change oil and filter
Clean crankcase breather
Change air cleaner element
Change coolant filter
Change fuel filters
Clean cooling system
Replacement of crankshaft
Change bearings
Change valves
Change valve springs
Change injectors
Change fuel pumps
Change piston
Change piston rings
Daily






Service time
Weekly Monthly 1/2 Year
Year




Estimated Cost
US $
35 to $ 70
70 to $ 140




140 to $ 280






280 to $ 560








35
1030 to $ 2060
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3.4.2 Cost of Diesel Generator
In Rwanda, the cost of available diesel generators is primarily dependent on the size of the generator
and the brand of the generator. For example, the smaller capacity generators have higher costs per
kW and the larger capacity generators have a lower cost per kW. Therefor the cost of diesel
generators does not vary linearly with the capacity of its output power.
The cost of diesel generator have been taken as shown in table 3.4 below [38] but the replacement
cost has been approximated depending with working aspect.
At present, in Rwanda the diesel price is 850 Frw ($ 1.2) and the transportation taken into
consideration since the diesel will be transported from the urban areas (Diesel Stations) to the rural
community [1]. Thus for this analysis the fuel cost has been considered to be $ 1.3.
Table 3.4 : Cost of Diesel generator on the market.
1
2
3
4
5
Size (kW)
10
15
20
25
30
Capital Cost($)
7000
8400
9800
10500
11200
Replacement ($)
5000
6000
7000
8000
9000
O&M Cost ($/hr)
0.5
0.6
0.7
0.8
0.9
3.5 Storage Battery
An off-grid hybrid system requires a storage to store the excess energy from the renewable sources
for later utilization when required. A backup in the system and the maintaining a constant voltage
during peak loads or a shortfall in generation capacity, batteries can help for the letter reasons [1].
Batteries are the most common storage method used in renewable energy system applications. A
battery is an electrochemical device capable to store energy in form of electricity when placing
different metals in an acid solution.
The Rwanda market we have two types of batteries; Primary batteries and Secondary batteries. The
batteries which can only be used only one time are called primary batteries while the batteries that
can be recharged are called secondary batteries. Renewable applications use secondary batteries.
The open circuit voltage of a battery is obtained by the type of electrodes and the electrolyte which
have been used. The voltage of a battery is not constant during charging and discharging of the
battery. Typically at the equilibrium conditions is known as the nominal battery voltage [4].
The battery chosen for this Study is 6CS25P-Surrette of 6V with a nominal capacity of 1,156 Ah
(6.94 kWh). The letter can be found on market with the capital cost of $ 1200, then the approximated
replacement cost and O&M costs for one unit of this battery has been considered to be, $ 1200 and
$ 30/year, respectively.
36
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 3.18 : Capacity curve of the Surrette 6CS25P, 6V battery, from Homer.
Figure 3.19 : Lifetime curve of the Surrette 6CS25P, 6V battery, from Homer.
The capacity of a battery is defined as the energy that can be withdrawn from starting to fullycharged state and it is measured in Ampere hours. But the capacity of a battery depends on the
proportion at which energy is withdrawn from it [1]. The higher the discharge current, the lower the
capacity. For example, the capacity curve of the Surrette 6CS25P battery is given in Figure 3.18.
The nominal capacity of this battery is given as 1156 Ah by the manufacturer and it is just one point
on this capacity curve.
The life of a battery primarily affected by the range of discharge and the temperature at which it is
operating. Depth of discharge is the level at which batteries are discharged in a cycle before they
are charged again. Usually the manufacturer specifies the nominal number of complete charge and
discharge cycles as a function of the depth of discharge in the product data sheet. For example,
Figure 3.19 gives the lifetime curve of the Surrette 6CS25P battery. The Figure indicates the number
of cycles to failure drops quickly with increasing depth of discharge [4]. It also shows the lifetime
of the battery which also depends on the number of cycles to failure [4].
37
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3.6 Inverter
An inverter is a vital element in any solar system where the AC is required. It converts the DC form
of solar system or wind system into AC form for AC appliances [1]. A hybrid system needs an
inverter to convert DC voltage from the batteries to AC voltage required by the load. Some aspect
is needed to be taken into consideration when selecting an inverter for a certain application. Usually
the inverters used in renewable applications can be divided into two; inverter for solar and for wind
electric system [39].
Inverters for solar electric system are also divided into four types depending with its application.
Stand-alone inverter or off-grid inverter, Grid connected inverter, hybrid power inverter and
Grid interactive inverter.
Also there are inverters which are specifically designed for PV applications and they are integrated
with Maximum Power Point Trackers (MPPT) and some inverters are bidirectional so that they can
operate in both inverting and rectifying modes. That is why a proper inverter must be carefully
selected for the power system according to the requirement and also paying attention to the hybrid
system configuration as well. With this project, we will use the following inverters:
Off-grid inverters, this is good for the remote stand-alone power system which has some batteries
for the backup. The pure sine inverters are the best for home and rural village systems.
Hybrid power inverters are good for the combination solar and diesel generator or any other RE
sources.
The DC side voltage of a battery inverter have to be matched with the battery bank voltage.
Generally stand-alone battery inverters operate at 12, 24, 48, 96, 120 or 240 V, DC depending on
the power level. For high power applications system it is required to use an inverter with a higher
DC voltage due to the current ratings of the wires and the rated capacities of other DC components
such as fuses, breakers decrease. The efficiency of the inverters have been improved a lot in this
days and 90 % and above is the typical efficiency. However, the latter varies with the load and
normally the manufacturer specifies the efficiency curve of the inverter.
The cost of the bidirectional inverter is $ 8239 for 10kW. By estimating the transportation and tax
all together on 38%, the capital cost comes to $ 11370. In this project the approximation of
replacement and O&M cost has been taken as $ 11370 and $ 2 /year respectively.
Table 3.5 : Inverter specifications [28][40][41].
Brand
Rated Capacity
Maximum Efficiency
DC voltage
AC voltage
Price
Solar Inverter
SMA Sunny TriPower
10 kW
98 %
330 V-800 V
230/400 V, 50 Hz
$ 4345
38
Battery Inverter
SMA
10 kW
95 %
41 V – 63 V
230/400 V, 50 Hz
$ 8239
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
4 Hybrid System Modelling
The chapter presents the modelling of hybrid system using the optimization software called
HOMER. The chapter start by describing the important inputs that demonstrate the technical
specifications, resources data and the costs which are relevant for modelling the system in HOMER
and the chapter will end by giving a brief discussion on how the software calculates the levelized
cost of energy using the economics inputs.
A brief summary of the site specific information that I entered into Homer
- Average load demand is 10 kW, peak load demand is 38 kW, and average of the daily
demand is 249 kWh/d and the load factor of 0.275.
- Water flow is 280 L/s and the head is 9.7 m.
- Solar irradiance is 5.13 kWh/m2/d, the clearness index is 0.513 and the average
temperatures is 22.7°C
- As explained in the section 2.4, that situation sets limits to this new hydro power stations.
The maximum water flow of 280 litres per second is a limitation to 20 kW as max power
from the micro hydropower plant.
- The battery is Surrette 6CS25P of 6V, 1,156 Ah (6.94 kWh)
- Diesel generator with a lifetime of 15,000 operating hours is used.
- Efficiency of 98 % for converter.
4.1
Introduction
The major components of the hybrid system with their technical details and relevant costs have been
discussed earlier in Chapter 3. As stated in problem statement the aim of this development is to find
out the best hybrid configuration which can supply the electricity at the lowest price with an
accepted level of availability. For this, it is required to consider several combinations of RES and
diesel generator with different component capacities. This is achieved by the software called
HOMER [1].
HOMER is a computer model that facilitate the assignment of assessing the design options for both
off-grid and grid connected hybrid systems for isolated, stand-alone and distributed generation
system. It facilitates a range of renewable energy and conventional technologies including solar PV,
wind turbine, hydro power, generator, battery bank and hydrogen. HOMER’s optimization analysis
algorithms help to assess the cost effective and technical practicability of a certain number of
technology options [4]. The sensitivity in HOMER allows to find the effect of uncertainty in the
input variables to the energy cost and the optimal configuration.
HOMER hybrid model requires several inputs which basically describe the technology options,
component costs, component specifications and resource availability. HOMER uses the energy
balance in optimization calculations. Electric and thermal load demand are compared to the
produced energy in that hour and then compute the flow of energy going to or coming from each
element of the system [4]. This comparison is done for every 8760 hours of the year for every
39
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
system type that the user wants to consider [4]. It then determines whether the hybrid configuration
can supply the demand under the conditions that the user has specified. If it can, then HOMER
calculates the NPV of installation and operating cost of the project throughout its lifetime and the
COE based on the Levelized (LCOE) [4]. The resulting hybrid configuration that has the least
LCOE or the least total NPV of the project is considered as the optimum hybrid system [4].
Figure 4.1 : Inputs required by HOMER hybrid model.
4.2 Modelling of Equipment
The hybrid system will be modelled in HOMER as shown in Figure 4.1 above by using AC coupled
hybrid configuration.
Once the required components of the hybrid system are selected as in Figure 4.1, there are several
inputs which must be entered in each of the component input windows in HOMER. These inputs
basically describe the costs, technical specifications and resource data that have been already
discussed in Chapter 2 and 3. A summary of these data are discussed here below.
4.2.1 Load input
Primary load is the one to be met immediately in such way that no unmet load [4]. An addition of
two separate primary loads to the system from the Add/Remove window. Each hour, HOMER
calculate the power produced by the elements of the system to serve the total primary load [4].
The baseline data is a number of 8,760 values that represent the average of electric demand and it
is expressed in kW, and this values is taken for each hour [4].
40
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Two technics to produce baseline data [4]: Either by HOMER to synthesize data, or by importing
hourly data from a file [4].
The technic of synthesize, just put at least one load profile, which is a set of 24 hourly values of
electric load [4]. 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.
HOMER adds randomness according to the values you enter for daily noise and hourly noise.
The daily and hourly noise inputs help to add randomness to the load data to make it more realistic
[4]. Figure 4.2 shows how this noises will affect the average load profile [4]:
Figure 4.2 : Random variability (daily and hourly noise) set to zero.
Firstly, take a look of load without any added noise [4]. As shown on the plot in Figure 4.3 below
of the first week of the year [4].
Figure 4.3 : Load plot without any added noise for the first week.
41
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
So without any noise, the load profile repeats precisely day after day [4]. In reality, the magnitude
and the form of the load will change from day to day [4]. This standard deviation come up with
more realistic load demand. With 10% daily noise and 10% hourly noise, a result of plot is shown
in the below Figure 4.4 for the first week in year one [4].
Figure 4.4 : Load plot with an added random variability for the first week.
Daily noise will affect the size but the shape doesn’t change where the hourly deviation affect the
shape but without affecting its size [4].
The mechanism for adding daily and hourly noise is simple. First HOMER bring together the 8760
hourly values of load data from the specified daily profiles [4]. Then it multiplies each hourly value
by a factor
α=1+δd+δh
(4.1)
Where; δd = daily perturbation factor and δh = hourly perturbation factor
The following Figure 4.5 is the results of different parameters which have been used and obtained
for this project from the homer load input window.
As decided in Chapter 2, a constant load profile has been assumed throughout the year, but hourly
and daily randomness has been added to this load profile in HOMER to generate realistic load
profile [17]. I have added 10 % randomness for both these cases. Adding 10 % randomness to the
load profile results increase in annual peak demand to 38.9 kW and the load factor of 0.275.
HOMER calculates the parameters, annual average of the daily demand, peak load and load factor
based on the load profile and the random variability inputs given by estimation.
42
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 4.5 : Homer primary load input window.
4.2.2 PV input
This opening is used to characterize the cost curve of PV panels, select the sizes that HOMER will
deal with for the optimal system, and specify the placement of the array [4].
In the cost table as shown in the Figure 4.6, the PV cost depend with how big is the system. Normally
this necessitate only a single row because PV costs are commonly assumed to be linear with size
[4]. In this project, the capital price of PV panels for a 10kW photovoltaics system has been
specified and taken at $35,000 and the cost for replacement is assumed to be $25,000. The O&M
cost is specified as $30.
The PV derating factor (fPV) [6]: This is a factor which is used by HOMER to consider some
factor that can affect the output power of PV in real-life operating conditions when relating with
the standard conditions [6].
It is used to take into consideration for those factors like dirtying of the panels, losses due to wires,
shadow, snow cover, oldness, and so on [6]. In this project, this factor have been estimated to 80%
so that Homer can calculate the output power by taking into account other factors which may reduce
the PV’s output power.
HEffect of Temperature on the PV Array: The PV Inputs opening gives the choice of explicitly
modelling the consequence of temperature exercises on the array [4][6]. The ambient temperature
43
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
will be required so that HOMER can use it in calculation of the cell temperature and this will be
executed in each time step [4][6]. If you desire to not explicitly consider the effect of it, then the
temperature-related difference between real and rated power output by reducing the PV derating
factor have still to be considered [4][6]. In this project, the choice of explicitly modelling the
temperature effect on array is selected and the ambient temperature data [4][6] in Table 2.3 from
NASA Surface Meteorology and Solar Energy website is used.
The slope (β) is the placement angle of the panels to the horizontal [4]. A slope of 0° means
horizontal, and 90° to vertical [4]. Taking the system as fixed-slope systems, an approximation of
slope equal to latitude of the location will practically maximize the PV energy production [4]. The
azimuth indicated the direction followed by the panels slope [4].
Therefore two axis solar tracker can be used to follow both of these sun’s movements. Generally
the costs of these two axis trackers are relatively high. Hence, they are not widely used in
commercial applications. In most the applications, the panels are mounted with a fixed slope.
Fixed slope solar collectors normally face towards the equator and the tilt angle is set to an angle
which is equal to the geographical latitude of the collector location on the earth. This angle is a good
to maximize the annual performance of the collector. In this project the slope would be 1.30 degree
of latitude, but this slope is very small by considering the rain water which may depose to the solar
modules, due to this an estimation of 15o have been set so that rain water may fall down very easily.
The azimuth (γ) is the orientation which is supposed to be followed by placement of panels in
terms of slope [4]. 0° for south, 90° for east, 90° for west and 180° for north [4]. If the azimuth
angle of the system is set to be fixed then the modules will be orientated as 0° azimuth for the
systems situated in the northern hemisphere and 180° azimuth for the systems situated in the
southern hemisphere)[4].
The ground reflectance (ρg) is the percentage of radiation that is reflected on the ground [4]. A
grass-covered is 20 %. Snow areas may go as high as 70 % [4]. For this project, the ground
reflectance is set to 20 %.
The temperature coefficient of power (αP) this express how the output power of an array depends
on its surface temperature [4][6]. Since output power is reduced with the increasing of cell
temperature, that is why this number is negative [4][6]. Manufacturers give this number in their
product brochures. In this project a monocrystalline solar panel SUNTECH STP250S has been
selected and in its data sheet, the manufacturer has specified the temperature coefficient of power
of -0.44 %/oC[27].
The nominal operating cell temperature (Tc,NOCT), this gives the level of how the solar radiation
and the ambient temperature affect the temperature of the PV array on its surface [4][6]. HOMER
uses this to compute the PV cell temperature [4][6].
PV manufacturers typically give this number in the data sheet [4][6]. The same case in this project
a crystalline solar panel SUNTECH STP250S has been selected and in its data sheet, the
manufacturer has specified the NOCT of 45oC[27].
PV Efficiency at Standard Test Conditions (ηmp,STC) [4][6]: HOMER uses this to compute the
PV cell temperature [4][6].
The following equation can be used to find out this number:
44
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
ηmp,STC = YPV/(APV*GT,STC) [6]
(4.2)
Where
ηmp,STC is the module efficiency tested under the standard conditions (%) [4]
YPV is the rated output power of the module tested under the standard conditions (kW) [4].
APV is surface area in m2 [4]
GT,STC is the radiation tested under the standard conditions (1 kW/m2) [4].
HOMER suppose the PV to work under maximum power point.
In this project a monocrystalline solar panel SUNTECH STP250S is selected and in its data sheet,
the ηmp,STC equal to 15.4% was stated by the manufacturer. The Figure 4.6 below summarize the
parameters required by homer for the PV input.
Figure 4.6 : PV input window, from homer.
4.2.3 Hydropower input
HOMER consider only a single size of hydropower system [4]. Reason being, no tables of costs or
sizes to consider in the opening for hydro Inputs [4]. Instead, it designate the cost and properties of
size of hydro to be considered [4].
Available Head (h), this is the difference in height between the intake and the turbine [4]. In this
project, the available head has been estimated to 9.7 m.
HOMER uses this number to compute the output power of the hydropower turbine [4].
45
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Effective Head (hnet); this is the actual head but minus the losses which have been take covet into
the penstock and it is expressed in term of head loss [4]. HOMER use equation 4.3 to compute this
net head [4]:
hnet =h*(1-fh)
(4.3)
Where
h = available head [m] and
fh = pipe head loss (%)
HOMER uses the effective head to calculate the power output of the hydro turbine in each time
step.
Pipe Head Loss (fh); Water and any other viscous fluid moving through a conduct encounter a loss
in pressure because of resistance in friction [4]. This number is specified in HOMER as a percentage
of the actual head [4].
The design flow rate (𝑄̇ design) is the rate of flow designed for hydro turbine [4]. It is the one which
operates the turbine at its maximum efficiency, although HOMER takes the latter efficiency as a
constant [4]. In this study, the flow rate has been taken as 280 liters per second.
The minimum flow rate, using the equation 4.4 can be calculated [4]. HOMER suppose that the
hydro turbine can run only if the flow is greater or equal to this minimum flow[4]
𝑄̇ min=wmin*𝑄̇ design
(4.4)
Where
wmin = The minimum flow ratio of the hydro turbine (%) [4]
𝑄̇ min = the minimum flow rate (m3/s) [4]
The maximum flow rate is the extreme flow that can be supported by the turbine [4]. To calculate
this flow, Homer use the equation 4.5 below[4]:
𝑄̇ max=wmax× 𝑄̇ design
(4.5)
Where
wmax: The maximum flow ratio of the hydro turbine (%)
𝑄̇ max: The maximum flow rate (m3/s)
The detailed parameters which have been discussed above can be summarized in the Figure 4.7 as
it is required by homer software to compute the output power of the desired micro hydro power
plant [4].
Economy
These inputs gives the details on system costs including the civil works and details on the turbine
resources [4] as shown in the Figure 4.7.
Capital cost; the initial cost for 20 kW hydropower system is taken as $40.000.
46
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Replacement cost is $30.000, O&M cost is taken as an approximation to $800.
Lifetime of the plant is 25 years.
Figure 4.7 : Hydro input window, from homer.
4.2.4 Diesel generator input
This opening in homer allows user to enter the cost, characteristics and performance of a diesel
generator [4].
Generator Minimum Load (fgen,min):This is the minimum percentage of rated capacity for the
generator for good functioning [4]. But this minimum load will not stop the generator from being
shut down, it will only avoid it from operating at too low load [4]. The existence of this parameter
is required because some manufacturers recommend that their machines not to be operated under a
certain load [4].
Lifetime (Operating hours of generator): This is the number of hours that can be used by the
generator during its life and the manufacturers of diesel generators usually provide this number of
working hours in their product brochures or data sheet. The lifetime of the selected generator has
been assumed to be 15,000 hours this project.
Generator Average Total Efficiency [4]: This is a summation of energy out; electrical and thermal
and divide the in energy, the latter is the energy of fuel [4]. The equation 4.6 is used to calculate the
efficiency [4][6]:
ηgen, tot =
3.6.(𝐸𝑔𝑒𝑛+𝐻𝑔𝑒𝑛)
𝑚𝑓𝑢𝑒𝑙.𝐿𝐻𝑉𝑓𝑒𝑙
Where
Egen is the electrical production per year (kWh/yr)
Hgen is the thermal production per year (kWh/yr)
mfuel is the used fuel per year (kg/yr)
LHVfuel is the lower heating value of the fuel (MJ/kg)
47
(4.6)
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
3.6 is because 1 kWh equal 3.6 MJ [4]
Electrical Efficiency, the homer use the equation 4.7 to compute it.
3.6 𝐸𝑔𝑒𝑛
ηgen = 𝑚𝑓𝑢𝑒𝑙.𝐿𝐻𝑉𝑓𝑒𝑙
(4.7)
Where the parameters are the same as in equation 4.6.
Economics
In the cost table as shown in the Figure 4.8, the cost changes with the power of the generator [4]. In
this project, a machine of 10 kW costs $ 7,000 initially, $ 5,000 for replacement and $ 0.50/h for
O&M [4]. A machine of 15 kW generator costs $ 8,400 initially, $ 6,000 for replacement and $
0.60/h for O&M [4].
The Figure 4.8 below shows the diesel generator input parameters required by the homer software
Figure 4.8 : Hydro input window, from homer.
4.2.5 Battery input
This window shown in Figure 4.9 allows the Homer user to select the type of battery, describe its
costs, and tell the software how many batteries to be considered for the optimal configuration [4].
Figure 4.9 : Batteries stored in homer component library.
This drop-down box contains all the batteries stored in homer component library. It help to choose
an appropriate battery model from this list. When selecting with this drop-down box, HOMER
48
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
shows more different types of batteries and its characteristics when you click on details button. In
this project a Surrette 6CS25P of 6V has been selected and 48V voltage system which require 8
batteries in series this means eight strings.
Battery Charge Efficiency: the software assumes the battery efficiency to be the square root of the
round trip efficiency [4]:
𝜂𝑏𝑎𝑡𝑡, 𝑐 = √𝜂𝑏𝑎𝑡𝑡, 𝑟𝑡
(4.8)
𝜂𝑏𝑎𝑡𝑡, 𝑑 = √𝜂𝑏𝑎𝑡𝑡, 𝑟𝑡
(4.9)
Where
ηbatt,c : charge efficiency,
ηbatt,d : discharge efficiency
ηbatt,rt : round trip efficiency[4].
Economics
In the figure below, this is the window that help to introduce the cost of battery [4]. I have entered
an amount of one battery for capital, replacement and O& M costs. In this project as shown on
below Figure 4.10, the costs of one unit is as follows $ 1,200 initially, $ 1,200 replacement, and $
30 annually for O&M [4]. If the number of batteries is n, then the costs for the total will be multiplied
accordingly [4]
Figure 4.10 : Battery input window, from homer.
49
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
4.2.6 Converter input
Any system that is consisted with both AC and DC configuration necessitate a converter [4]. This
opening help to introduce the cost curve for the converter and preferred sizes to be for the optimal
system [4].
Lifetime: operating years [4]. In this project 15 years has been assumed.
Efficiency: The efficiency for converting DC form to AC form [4]. Homer will use in the
calculations the efficiencies which have been entered in the its input data. In this project the
efficiency from the manufacturer of the inverter have been taken as 98 %.
The efficiency of rectifier to convert AC form to DC form [4], in % is taken as 95.
Economics
In the cost table as shown in the Figure 4.11, the cost of converter depend with its size. In this
project, the capital and replacement price for 10 kW converter is taken at $ 11,370.
Figure 4.11 : Battery input window, from homer.
50
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
4.3 Modelling of Resources
In HOMER, a "resource" is those external parameters used by component in order to have electric
or thermal energy [4]. Hydro, solar radiation, wind and diesel fuel are resources [4].
4.3.1 Solar resource inputs
This window is used to specify the latitude, longitude and the solar radiation of the selected area for
the photovoltaic system in the year [4][6]. HOMER uses these data to compute the hourly output
power in the year. HOMER can also calculate hourly solar radiation data based on the monthly
average radiation and clearness index. There is a method which is used to produce synthetic hourly
solar radiation data. Figure 4.12 shows the synthetic in Burera which is calculated by HOMER
using the NASA radiation data.
Figure 4.12 : Synthetic solar radiation data over a period of a year.
The latitude specifies the location of selected community on the Earth's surface. It is an important
variable in solar calculations. This parameter is required for calculating radiation values from
clearness indices, and conversely. To calculate the radiation incident on a tilted surface that one can
also be used. In this research, a village located at 1o30’ S latitude and 29o58’ E longitude is selected
for studying the hybrid system.
This results has been presented in the Figure 4.13.
51
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 4.13 : Solar resource inputs window, from Homer.
4.3.2 Hydro Resource inputs
Hydro Resource opening is used to describe and introduce the flow available to the hydro system
[4]. The HOMER software will use these parameter to computer the annual output power of the
hydropower plant [4].
The set of 8,760 called baseline values characterizing the average flow in L/s, for each hour of the
year [4].
The residual flow is the unused water for ecological reasons for fish and other aquatics animals
[4]. HOMER uses this number for computing the available flow rate to the hydro turbine [4].
52
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 4.14 : Hydro resource inputs window, from Homer.
4.4 Modelling of Other Important Factor
4.4.1 Search space window
This window gives convenient access to the search space, which is the set of all allowable sizes and
quantities of each component. You can also specify this information separately in each of the
component input windows (for example, you can enter the allowable sizes of the PV array in the
PV inputs window) but the Search Space window allows convenient access to the entire search
space. The search space for this project is presented in the Figure 4.15.
In its optimization procedure, HOMER evaluates each configuration type and then grade each
according to the total NPC. The specification of the search space almost always involves a tradeoff between accuracy and run time.
53
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 4.15 : Values of elements optimization.
4.4.2 Economic inputs
The interest rate also known as discount rate. It is used to convert between one-time costs and
annualized costs [4]. It is seen in the Economic window [4]. In order to evaluate the economic
viability of micro-grids supplied by hybrid power systems, the levelized cost analysis of systems
must be done by considering the lifetime of the system, because hybrid systems with renewable
technologies have a higher capital cost, but small O & M cost during the life of the system [1]. In
contrast, fossil fuel based electricity generation systems have lower capital cost but higher O & M
cost due to fuel costs, generator maintenance and replacement costs. Thus, LCOE analysis can
compare the economics of different technological solutions.
HOMER optimization algorithms are based on the LCOE analysis. It finds the optimum system by
calculating the Net Present Value (NPV) of the lifetime cost of the project by including all the costs
that arise within the life of the project for every system configurations considered in the search
space [4]. Then it grades all the possible configuration types according to increasing NPC and
LCOE [4]. The following gives a brief summary on the total NPC and the LCOE [4].
The total Net Present Cost
The total Net Present Cost (NPC) of the system is the difference between the present values of all
the costs occurs over the project lifetime and the present values of all the revenue earns over project
lifetime [1][4]. The present value of the costs that will make n-year later can be calculated by the
following formula.
𝐶𝑁𝑃𝐶 = 𝐶(
1+𝑖′ 𝑛
)
1+𝑑
(4.10)
Where
𝑖 ′ : is the annual inflation rate (%)
d: is the nominal interest rate (%)
54
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
In order to find the LCOE, the total NPC of the project must be converted to series of equal annual
cash flows which is known as total annualized cost calculated by the equation 4.11 [4].
Total annualized cost ($/year) = Total NPC*CPF
(4.11)
Where, CPF is the capital recovery factor and it is given by the formula,
𝑖(1+𝑖)𝑁
CPF = (1+𝑖)𝑁 −1
(4.12)
Where
N: number of years
i is the real interest rate,
The lifetime of the project has been studied here is 20 years. Therefore N is 20. The real interest
rate is determined using the nominal interest rate (d) and the annual inflation rate (𝑖 ′ ) by equation
4.9 [4].
𝑖=
𝑑−𝑖′
(4.13)
1+𝑖′
HOMER suppose the rate of inflation to be the same for all types of costs (fuel cost, maintenance
cost, labour cost, etc.) occurring over the life of the project [4]. The variation of the real interest in
Rwanda over the years is given in Figure 4.16. According to Figure 4.16 the average real interest
rate in Rwanda during the past 32 years (1978 – 2010) has been about 8.4 % [42]. If the effect of
the peak occurred in 1986, 1999 & 2002 is removed, the real interest become 6.65 % is selected for
the analysis and it is presented in Figure 4.17.
Real interest rate (%) in Rwanda over the past 32 years
35
Real interest rate %
30
25
20
15
10
5
0
1975
-5
-10
1980
1985
1990
1995
2000
2005
2010
2015
Year
Figure 4.16 : Changes in the real interest rate in Rwanda over the past 32 years [42].
55
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Levelized Cost of Energy (LCOE)
It is the cost per kWh of electrical energy, such that the total NPC of the useful energy generated
throughout the whole lifetime of the hybrid project is equal to the total NPC of the project [4]. The
calculation of LCOE of the electricity generated by an off grid hybrid system is done as shown in
the equation 4.14 [4].
𝐿𝐶𝑂𝐸 =
𝑇𝑜𝑡𝑎𝑙 𝐴𝑛𝑛𝑢𝑎𝑙𝑖𝑧𝑒𝑑 𝐶𝑜𝑠𝑡 (𝑈𝑆𝐷/𝑦𝑟)
𝐴𝑛𝑛𝑢𝑎𝑙 𝑙𝑜𝑎𝑑 𝑠𝑒𝑟𝑣𝑒𝑑 (𝑘𝑊ℎ/𝑦𝑟)
(4.14)
System fixed capital cost
The fixed capital cost of the system is normally allocated for building a house for keeping the battery
bank, charge controllers, generator, inverter and other relevant electrical instruments and
constructing the distribution lines throughout the village. It also includes the site preparation cost,
labour cost, engineering design cost and other various cost. An estimate of the fixed capital cost by
considering 3 km long three phase distribution lines has been estimated as $ 45,000.
System fixed operation and maintenance cost
System fixed O & M cost firstly includes labour cost and insurance costs. If a full time engineer or
a technician is working in the hybrid system place, then he has to be paid a monthly salary.
Assuming a technician is full time employing in the power house annual fixed O & M cost has been
taken as $ 7,000.
The project lifetime
It is the number of years the project is forecasted in the [4]. HOMER uses this number to compute
the annualized replacement cost and annualized capital cost of each component, as well as the total
NPC of the system[4]. In this project, the project lifetime has been taken as 20 years.
Figure 4.17 : Economic input window.
56
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
4.4.3 System control inputs
The diesel generator is the only none renewable and dispatchable energy source used in this hybrid
energy system. Since the energy output from renewable sources is highly intermittent and cannot
be controlled by the user, then it have to be used when it is available for supplying the load or to
charge the battery bank. If renewable energy systems or the battery bank is not able to meet the
load then the diesel generator has to be auto-started so that it can supply the load without causing
power interruptions. Therefore using a diesel generator is essential in hybrid systems to supply the
load in a controlled mode to improve the availability of the system. However controlling the
performance of a diesel generator is rather complex due to several aspects. The main aspect is the
efficiency conversion of energy. As it is discussed in chapter three, the generator’s efficiency is
very low at low load factors. Therefore if the generator is chosen to supply the load that cannot be
supplied by the renewable sources, then the diesel generator operate in a low load factor which lead
to a low efficiency. Normally, operating the generator at its full capacity when required and using
the excess energy for charging the battery bank may be more economical than the previous case.
This one is optimal approach which depends on many factors; like the power size of the generator
[17]. Therefore, dispatch strategy also should be developed when optimizing the hybrid system.
Two dispatch types [4]; i.e. “Load following” and “Cycle charging”.


Load Following (LF): The diesel generator is switched on when required and produce only
the required amount of power that cannot be produced by the renewable sources or battery
bank to supply the load.
Cycle Discharging (CC): The diesel generator starts when required and operates at its full
capacity and excess energy is sent to the battery bank to charge the batteries.
4.4.4 Component costs summary
Table 4.1 : The summary of the costs of components and other relevant costs.
Component
PV system
Hydro system
Battery
Generator
Inverter/charger
Capacity
10kW
20kW
1156Ah
10kW
15kW
20kW
25kW
29kW
10kW
Capital cost $
35000
40000
1200
7000
8400
9800
10500
11200
11370
57
Replacement Cost $
25000
30000
1200
5000
6000
7000
8000
9000
11370
O & M Cost $
30/ year
800/ year
30/ year
0.5/ year
0.6/ year
0.7/ year
0.8/ year
0.9/ year
5/ year
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
5 Results
The optimal hybrid system is the one which can supply electricity needs at the lowest price or in
other words, the system which is having the lowest total net present value, while supplying the
electricity at the required level of availability. In this chapter we will discuss the results obtained
from the HOMER simulations and the selection of the optimal system based on the simulation
results. The chapter also discusses the performance of the optimal hybrid system, hybrid system
design, economic viability of the project and a brief introduction to the energy management in the
micro grid.
5.1 Optimization Results
For the off-grid electrification of a village in Burera, various combinations of hybrid Systems have
been obtained with SPV, MHP, DG, batteries and convertors from the HOMER optimisation
simulation. Figure 5.1 presents a screen shot containing the summary of simulation outcomes.
All possible hybrid system configurations are listed in ascending order of their Total net present
cost. The best possible combination of MHP, DG, Batteries and Convertor is highlighted in blue,
and the next best possible combination, is highlighted in yellow, includes the SHP, Batteries and
Converter. The blue highlighted combination is able to fully meet Burera’s load demands at the
lowest possible total net present cost.
The optimal system configuration for our case study is 20kW
MHP, 10kW DG, 8 Surrette
6CS25P batteries and 10kW Inverter with a dispatch strategy of load following.
There is no PV selected at this site as it can be seen on Figure 5.1.
As indicated in Figure 2.4 of the daily village load profile, the maximum demand will take cover in
between 18h 00 and 21h 00 which is around 28 kW. But this power is changed to 38 kW due the
random variability of 10 for both day to day and time step to time step as described in Figure 4.5.
That is why, the system requires more source than the 20 kW from the micro hydropower plant.
The reason behind, it is not feasible to increase the output power generated by this micro
hydropower plant due to the limitation of design flow rate which cannot go beyond 280 litres per
second. So, an additional sources to generate the excess of 18 kW which cannot be produced by the
MHHP are required. The diesel generator and the batteries as shown in Figure 5.1 is optimized by
the homer software. As it has been discussed in section 3.4 of chapter 3 that the diesel generators
operate most efficiently when working near its full load, that is why, the diesel generator and the
batteries. These two sources is used to generate the extra power of 18 kW required during the peak
hours. The production Figure plots for the explanations of this configuration can be seen in the
appendix D.
This system is considered at $ 1.3/l of diesel cost and 280 l/s of design flow rate for the MHP. The
total net present cost, capital cost and the COE for such a hybrid system are $ 199,231, $ 112,970
and $ 0.201/kWh, respectively.
58
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
The COE of $0.201/kWh from this hybrid system is not cheaper than that of $ 0.2/kWh from grid
connected as considered for this study.
Figure 5.1 : Summary of HOMER optimization results in categorized way.
Table 5.1 : Optimal least cost hybrid system for the case study.
Cost summary
Total NPC
$199,231
LCOE
$0.201/kWh
Operating
$7922
Cost
Capital Cost $112,970
System architecture
MHHP
20kW
DG
10kW
Battery bank
8 Batteries
Inverter
10kW
Dispatch Str. Load Following
Renewable fraction 0.996
Capacity shortage
0
Electrical
Component kWh/y
%
Hydro
Turbine
DG
198,086 100
0
Total
198,929 100
843
Figure 5.2 : Electricity production from the best system type.
As shown in the results from the above Figure 5.1, a micro hydropower and a diesel generator only
system requires a small capital investment ($ 93,400) compared to hybrid systems. But due to the
large O&M cost due in general to a diesel generator ($10,851), the lifetime NPC of the system ($
211,549) is very much higher than the hybrid system ($ 199,231). Thus the LCOE is also becoming
larger which is approximately 0.214 $/kWh.
59
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Figure 5.3 : Optimization results when using only renewable resources.
According to the above observations we can see that the hybrid system which use both renewable
sources and diesel generator is more economical than the system with only renewable energy
sources, the letter is shown in the above Figure 5.3 where the optimal system configuration is 20kW
MHP, 16 Surrette 6CS25P batteries and 15 kW Inverter with a dispatch strategy of cycle
discharging.
But the important matter is the sizing of the hybrid components in a right way to reduce the energy
cost or net present cost of the project. If sizing is not done properly then we may end up with a
system having larger NPC than the base system was in concern.
The Figure 5.4 and Table 5.2 shows the cost flow summary for this project optimal system. The
Capital cost of micro hydro power plant is about 35 % and the fixed capital cost for the system is
about 40 % and the remaining share the capital cost of 25%.
About 84 % of the operation and maintenance cost goes to the monthly salary of an engineer charged
to run and control the system.
Figure 5.4 : Cost flow summary by cost type
60
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Table 5.2 : Cost summary of the project based on the used component.
Component Capital ($)
Hydro
40,000
Generator
7,000
Battery
9,600
Converter
11,370
Other
45,000
System
112,970
Replacement ($) O&M ($)
0
8,711
0
1,127
4,433
2,613
4,329
22
0
65,331
8,762
77,803
Fuel ($)
0
5,325
0
0
0
5,325
Salvage ($)
-1,656
-999
-883
-2,091
0
-5,629
Total ($)
47,055
12,453
15,764
13,629
110,331
199,231
These costs are further illustrated in Figure 5.5 as a nominal cash flow of the project throughout 20
years. Here we can see, after 11 years battery replacement occurs, the same case for the converters,
the letter will be replaced after 14 years. But the diesel generator replacement does not occur within
20 years lifetime of the project, because the hours of working for the generator during a year is
about 207 this means the total working hours of the machine during the project lifetime is 4140
hrs which is less than the lifetime operating hours of the generator 15000 hrs.
Figure 5.5 : Nominal cash flow of the project throughout 20 years.
Figure 5.6 shows the grid extension cost compared with the cost of a standalone system, it is clear
that this standalone system will be more useful for a community situated at a distance greater than
8.3 km from the national grid, otherwise it will be better to use the extension of national grid because
it is the one which will have less investment in terms of money. This decentralised RE hybrid system
is a better option than the grid extension for a selected community, because this village is at 20 km
from the transmission line of the national grid.
61
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Electrification Cost
250,000
Grid extension
Standalone system
Total Net Present Cost ($)
200,000
150,000
100,000
50,000
0
0
3
6
Grid Extension Distance (km)
9
12
Figure 5.6 : Breakeven grid extension distance with its cost
The optimal hybrid system configuration obtained from the HOMER optimization algorithm is
given in Table 5.1. However, finding the optimum hybrid configuration requires more analysis
because the inputs given to the HOMER may not 100 % accurate. Especially the renewable resource
inputs, solar radiation and water flow data. In this study these data have been obtained from the
NASA surface meteorology and solar database. Therefor there may be little variations between
these and the actual data. Moreover, there may be uncertainty in the cost inputs as well. Thus the
sensitivity analysis is required to validate the results obtained from the optimization analysis by
considering the sensitivities of the input variables.
5.2 Sensitivity Results
HOMER sensitivity algorithms is used to assess the effect of uncertainties in the input variables
discussed in section chapter 4, in selecting the optimum hybrid system configuration.
Sensitivity analysis exclude all impractical configuration and ranks the feasible configuration by
judging uncertainty of parameters [1]. HOMER allows taking into account future developments,
such as increasing or decreasing load demand as well as changes regarding the resources, for
example fluctuations in the river’s water flow rate, wind speed variations or the biodiesel prices.
Here, various sensitive variables are considered to select the best suited combination for the hybrid
system to serve the load demand. I have analysed the uncertainties of the following variables when
selecting the optimal hybrid configuration.

Design flow rate
It may have some fluctuation of water flow rate in the Rugezi catchment which will reduce
the water in the reservoir shown in Figure 2.7 where the power plant is possible [1]. That is
62
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda

why the sensitivity analysis has been done for the uncertainties of water flow rate in between
200 to 300 litres per second.
Diesel price
As there can be uncertainties in the estimated diesel cost of our generator, the sensitivity
analysis has been done for the uncertainties of diesel cost in between $ 1.1 to $ 1.5 per litre.
The techno economic optimum configuration system capable to supply power to a community
selected in Burera district in the northern province of Rwanda which its load profile has been
approximated and indicated in Figure 2.4, has been found using the HOMER optimization analysis.
The architecture of this system is given in Table 5.1.
As can be seen in the HOMER optimization and sensitivity analysis results given in Figure 5.7, both
capital cost, total NPC and LCOE can be affected with the design flow rate and the diesel price.
With the results of optimization and sensitivity analysis shown in Figure 5.7, the optimum
configuration has the total net present cost, capital cost and the COE for such a hybrid system are $
196,502, $ 112,970 and $ 0.2/kWh, respectively.
The COE of $0.2/kWh from this hybrid system is the same as that one from the national grid. But
there is no difference between the energy costs of this system configuration and the system
configuration ranked in the 1st place in the table 5.1 without sensitivity.
Figure 5.7 : HOMER optimization and sensitivity results in categorized way
The surface plot for the levelised COE is presented in Figure 5.8. The micro hydropower design
flow rate is represented on the x-axis, and diesel price variation on the y-axis. In Figure 5.7 as the
design flow rate increases, the power output from MHHP increases and consequently there is a
reduction in the total NPC. As the total NPC reduces, the system’s COE decreases as well. The
same case happen for the diesel price but in reverse way, the total NPC increases in little way as
well as the diesel price increases. Therefore, a hybrid system with MHHP proves to be the cheapest
option compared to other RETs due to lower capital cost of a small hydropower plant.
63
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Levelized Cost of Energy
1.5
226,054
220,955
213,930
Legend
206,250
200,051
0.2300 $/kWh
196,502
0.2265
0.2230
0.2195
0.2160
1.4
Diesel Price ($/L)
225,459
220,635
212,771
205,506
199,641
196,283
0.2125
0.2090
0.2055
0.2020
1.3
224,865
220,315
211,613
204,762
199,231
0.1985
196,064
0.1950
Superimposed
Total NPC ($)
1.2
224,271
219,995
210,454
204,019
198,822
195,845
223,677
219,675
209,296
203,275
198,412
195,626
220
240
260
280
1.1
200
300
Design Flow Rate (L/s)
Figure 5.8 : Surface plot of cost of electricity from hybrid system.
Figure 5.9 shows the result for the breakeven distance for grid extension or Economical Distance
Limit (EDL). It shows that the distance varies from 7.5 km to 11 km depending on the total NPC
and levelised COE. For the hybrid configuration for this study as shown in Figure 5.9, the total NPC
line comes out be a lower value than the breakeven distance for grid extension at a distance of 10.2
km, meaning that a decentralised RE hybrid system is a better option than the grid extension for a
community village which is at a distance greater than 10.2 km far from the national grid. It is clearly
evident from the line graph that as the design flow rate increases with a diesel cost at a fixed value
of $ 1.3/L, the total NPC of the system decreases. At 225 L/s of design flow rate the EDL comes
out to be 10.2 km, and at 300L/s the EDL comes out to be 7.9 km of distance. Hence, the total NPC
and levelised COE of a system determine the EDL with respect to the input parameters.
Total Net Present Cost vs. Design Flow Rate
225,000
11.0
215,000
9.5
210,000
9.0
205,000
8.5
200,000
Breakeven Grid Ext. Dist. (km)
Diesel Price = $1.3/L
10.0
Total Net Present Cost ($)
Fixed
10.5
220,000
195,000
200
Total Net Present Cost
Breakeven Grid Ext. Dist.
8.0
220
240
260
280
7.5
300
Design Flow Rate (L/s)
Figure 5.9 : Line graph for total NPC vs. design flow rate and breakeven grid extension distance
The Figure 5.10 which shows how the number of batteries can be affected by the design flow rate,
it is clear that the batteries increases with the increase of water flow rate, but the variation of fuel
price do not affect the battery bank for the hybrid system. The same case happen for the Converter
capacity as shown in Figure 5.11 this capacity of converter increases with the increase of water flow
rate and this capacity do not vary with the variation of fuel price.
64
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Number of Batteries
1.5
16
16
Legend
8
8
8
16.0
8
15.2
14.4
13.6
12.8
1.4
Diesel Price ($/L)
16
16
8
8
8
8
12.0
11.2
10.4
9.6
1.3
16
16
8
8
8
8.8
8
8.0
Superimposed
Number of Batteries
1.2
16
16
1.1
200
16
8
8
8
8
8
16
8
8
8
220
240
260
280
300
Design Flow Rate (L/s)
Figure 5.10 : Number of batteries vs the water flow rate.
Converter Capacity
1.5
15
15
Legend
10
10
10
15.0 kW
10
14.5
14.0
13.5
13.0
1.4
Diesel Price ($/L)
15
15
10
10
10
10
12.5
12.0
11.5
11.0
1.3
15
15
10
10
10
10.5
10
10.0
Superimposed
Converter Capacity (kW)
1.2
15
15
1.1
200
15
10
10
10
10
10
15
10
10
10
220
240
260
280
300
Design Flow Rate (L/s)
Figure 5.11 : Converter capacity with respect to the water flow rate.
Breakeven Grid Ext. Dist.
1.5
11.0
10.5
9.8
Legend
9.0
8.3
12.0 km
8.0
11.5
11.0
10.5
10.0
1.4
Diesel Price ($/L)
11.0
10.5
9.7
8.9
8.3
8.0
9.5
9.0
8.5
8.0
1.3
10.9
10.4
9.5
8.8
8.3
7.5
7.9
7.0
Superimposed
Breakeven Grid Ext. Dist. (km)
1.2
10.8
10.4
10.8
10.4
9.3
220
240
1.1
200
9.4
8.8
8.2
7.9
8.7
8.2
7.9
260
280
300
Design Flow Rate (L/s)
Figure 5.12 : Breakeven grid extension distance with respect to hybrid system
According to Figure 5.13 we can see that the LCOE decreases as the design flow rate increases. But
for the case of diesel price as it shown in Figure 5.14, the COE increases due to increase in the fuel
65
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
cost [15]. The decrease in the COE for the variation in design flow in the range of 200 Litres – 300
Litres is approximately $ 0.035/kWh (25 frw/kWh). But the change in the cost is not as high as in
the case of fuel price variation. It is not significant, see the Figure 5.14. The increase in COE for
the variation in diesel price in the range of $ 1.1 – $ 1.5 is approximately $ 0.0017/kWh (1.2
frw/kWh) which is negligible.
Levelized Cost of Energy vs. Design Flow Rate
0.230
Fixed
Diesel Price = $1.3/L
Levelized Cost of Energy ($/kWh)
0.225
0.220
0.215
0.210
0.205
0.200
0.195
200
220
240
260
280
300
Design Flow Rate (L/s)
Figure 5.13 : LCOE at different design flow rate.
Levelized Cost of Energy vs. Diesel Price
0.2025
Fixed
Design Flow Rate = 280 L/s
Levelized Cost of Energy ($/kWh)
0.2020
0.2015
0.2010
0.2005
0.2000
1.1
1.2
1.3
Diesel Price ($/L)
1.4
1.5
Figure 5.14 : LCOE at different diesel price.
5.3 Futures Connection of the Hybrid System to the National Grid
The selected rural community in Burera district in the northern province of Rwanda for this project
has not yet been electrified by the national grid. The government of Rwanda has a target of having
the electrification of 70 % of its citizens connected to the national grid. That is why, I assume that
the selected community will be included so that in the future the government may invest money for
the extension of the national grid to this remote village as well. But the time that will be taken to
accomplish this task is uncertain.
However, if the grid is extended to the selected rural village during the lifetime of the hybrid system,
then it will affect the cost returns of the project. It can influence either positively or negatively,
66
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
because once the grid is available, the community will buy electricity from the national grid. On the
other hand all the energy generated by the hybrid system can be sold to the grid. Therefore it is
important to analyze the cost returns of the project if the national grid is available in the future.
For this analysis I have made the assumptions that the community will have access to the electricity
from the national grid after any time and the hybrid power system is developed and operated by an
Independent Power Producer (IPP). Under these assumptions, the community will buy electricity
from the micro grid based on RET or in other words from the Independent Power Producer during
the first time and then from the national grid. Hence the IPP can sell the electricity generated by the
hybrid power system to the community during the period where national grid is not available. Once
the national grid is available, the hybrid system can be connected to the grid and all the energy
generated by the hybrid system can be sold to the national grid. When the hybrid system is
interconnected to the national grid, the generator and the battery bank will not be required any more,
thus they can be decommissioned. Then the hybrid system will consist only renewable system that
is, a MHPP with the capacity of 20 kW.
From the homer simulation results, the LCOE for the hybrid system is $ 0.201 and the COE from
national grid is $ 0.2, the difference in between to energy cost is $ 0.001 (0.7 frw/kWh) which is
negligible. From the cost above, it is clear that the project will run and earn the cost returns of the
project as it has been designed, but by selling all the energy generated by the hybrid system to the
national grid, this will benefit to the IPP because even the excess of electricity will be purchased,
which was not the case for the off grid system.
5.4 Design of the Hybrid System
According to the HOMER simulations it has been found that the following hybrid configuration is
the optimal hybrid configuration which can supply the electricity at a lowest cost with an accepted
level of availability to the selected one village from Burera District in the northern province of
Rwanda.
Micro Hydropower plant capacity
20 kW
Diesel Generator capacity
10 kW
Battery bank/Number of 1156 Ah batteries
55.52 kWh/8
Capacity of the bi-directional inverter
10 kW
The connection diagram of components is indicated in Figure 5.15. An AC coupled hybrid
configuration is used in designing the system. In this connection, all sources are connected to the
AC bus. As the batteries generate DC electricity, they are connected to the AC bus via bi-directional
inverter. Since the generator and the alternator from the micro hydropower plant give the AC power,
they are connected directly to the AC, see Figure 5.15.
A power house for micro hydropower plant will be used with all necessary equipment. Like;
Hydropower Gilkes Turgo turbine of 20kW, the butterfly valve motor drive unit, laptop for
monitoring the top of falls water intakes, the Dump load controller to dumps the excess of
electricity, Ethernet/communication box, capacitor bank for power factor correction and
proportional valve control.
The battery bank consists of 8 Surrette 1156 Ah, 6 V flooded lead acid deep cycle batteries
connected in series. The selected voltage of the DC bus is 48 V. Therefore 8 batteries have been
67
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
connected in series to obtain the nominal voltage of the DC bus of 48 V. one set of series connected
batteries resulting in nominal capacity of the battery bank of 55.5 kWh.
A one SMA Sunny TriPower Island–10 kW inverters have been used as the bidirectional inverter.
Village Load
230/400V,50Hz
MCCB
MCCB
Micro Hydropower Plant with
Gilkes Turgo 20 kW
BDG12LS 10 kW Diesel Generator
Battery Fuse
SMA Sunny TriPower bi-directional
inverter 10 kW
AC Conductor
DC Conductor
The battery bank of 8 Surrette
6CS25P 1156 Ah, 6 V in Series
Figure 5.15 : Single line diagram of the hybrid system
Fuses, Miniature Circuit Breakers (MCB) and a Moulded Case Circuit Breaker (MCCB) are
employed in the system to protect the equipment and conductors from overload and over current
conditions. A MCCB is used to limit the total load of the micro grid to the required level and also
to protect the equipment from over currents created due to faults in the micro grid. Since the
maximum power of the village is about 38 kW. MCCB with a nominal rated capacity of 60 A is
used. The thermal release setting (for overload) and magnetic release setting (for over current) can
be set with respect to the nominal rated capacity in the MCCB.
In order to maintain the proper operation of the hybrid system by maximizing the energy utilization
generated by the renewable sources and minimizing the operating hours and fuel consumption by
the diesel generator, the collective operation of the individual components should be controlled by
an intelligent hybrid management system. Energy Management System (EMS) can be either
automatically or manually operated. A manually operated system requires well trained operator for
68
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
controlling the system operation. On the other hand automatic system does not require human
involvement. However, the design of the energy management system of the hybrid system is not in
the scope of this thesis.
5.5 Economic Viability
Development of a rural electrification scheme based on a renewable hybrid power system in Burera
district requires an initial capital investment of approximately $ 113,000. This system can feed
approximately 150 households including public utilities and several small businesses. This type of
rural electrification schemes can be implemented either private sector based, utility based or as a
combination of private sector and utility based. However, the government contribution will be
essential to make the service affordable to the end users and to ensure the sustainability of the
system even though the project is developed in either way, because the levelized cost of electricity
(0.201 $/kWh) is basically equal to the average price of the electricity from the national grid. In
reality, the electricity price of rural electrification schemes cannot be equalized with the national
grid tariff which already incorporates subsidies, particularly in developing countries. Nevertheless,
development of proper tariff structure is a crucial factor in the design stage of the project in order
to attract the private sector investors. Simultaneously, sustainable financing for O & M of the
system must be ensured by the regulated purchase tariff.
Generally the basic rule in the electrification project is that the tariff structure must cover at least
the capital and the lifetime O & M cost of the project [4]. The LCOE is the indicator that represents
the flat electricity tariff that can cover the capital and the O & M cost of the project during the
project lifetime [4]. Table 5.3 illustrates how the levelized cost of energy can be brought down with
the capital donation from the government or Non-Governmental Organizations (NGO). Other than
capital investment based subsidy, several subsidy schemes are available. Different schemes can
lower the energy cost by different amounts. As showed in Table 5.3, the LCOE can be reduced by
0.133 $/kWh if subsidy is available for covering 60 % of the capital cost.
Table 5.3 : Effect of subsidies on the electricity price.
Donation as percentage of capital cost (%)
LCOE ($/kWh)
0
0.201
15
0.184
30
0.167
45
0.15
60
0.133
In contrast to the financial scheme, proper O & M scheme must be developed to ensure the
sustainable operation of the rural electrification project. Local people can be trained for doing basic
maintenance of the system and even for collecting the monthly fees from the consumers. O
& M cost can be brought down by incorporating well trained local people. However, the service of
skilled technicians will be required for major maintenance activities especially in the micro
hydropower plant for turbines and alternator and the diesel generator. Table 5.4 illustrates how the
costs can be brought down by involving local trained people for maintaining the system. Training
local people for monitoring the system operation and doing routine maintenance of the system will
avoid the necessity of employing a full time technician thus reduces the administrative cost.
69
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Table 5.4 : Effect of system fixed O & M cost on the electricity price.
System fixed O&M cost ($/year)
LCOE ($/kWh)
6,000
0.201
5,000
0.19
4,000
0.179
3,000
0.168
2,000
0.157
1000
0.146
If the annual fixed O & M cost can be brought down approximately to $ 1000 then the energy cost
drops from 0.201 to 0.146 $/kWh, and with the capital subsidies of 60 %, the energy cost can be
brought down approximately from 0.201 to 0.133 $/kWh. This is an acceptable and affordable
price for the rural consumers.
5.6 Efficient Use of Electricity in the Micro grid
Efficient use of electricity is a key issue in rural electrification systems which are based on the
hybrid power systems, because, the rated capacity of the generating systems is increasing as the
peak demand of the community increases which then increase the capital cost substantially. To
reduce the capital investment, the peak and the average load must be minimized as much as could
be [1]. Therefore, the inhabitants must be educated about the energy efficiency measures. At the
same time, people should aware about their responsibilities and the system limitations. Usually in
rural communities, lighting loads make the largest contribution to the total load. Thus people should
be educated about the energy efficient bulbs (LED) and their long lifetime to eliminate purchasing
cheap low efficient incandescent bulbs. In addition people should be given an understanding about
the right time of using the certain electrical appliances such as electric irons and water pumps. By
operating them during off peak hours the peak demand can be reduced. In addition, people must be
encouraged to iron cloths of few days and avoid daily ironing. Further, suitable water pumps must
be selected according to the pump lift and flow rate. All these can be done by conducting awareness
programs for the inhabitant in the village. So, awareness programs also should be a part of the micro
grid development scheme.
On the other hand an alarm system can be designed to avoid unwanted tripping of the main circuit
breaker due to increase in load above the rated capacity of the hybrid system. For this, the village
load should be continuously monitored and as it is about to reach the limits, a siren horn can be
activated to encourage the people to switch off some electrical appliances for clipping the peak load.
The people may give positive feedbacks on these alarms, because they know otherwise they have
to face supply interruptions.
5.7 Comparison of Electricity Prices
Solar Home Systems (SHS) provide huge benefits to the people in rural communities where national
grid electricity is not available. These products are accessible from 10 Wp – 100 Wp and the prices
are proportional to the size and the complexity of the system [1]. However the costs of these systems
are typically higher and cannot be afforded by low income rural inhabitants for a one-time payment.
Therefore, several loan schemes are available to make these systems viable for rural low income
70
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
families in Rwanda. The cost of energy generated by these systems typically lies in the range of
0.28 – 0.30 $/kWh which is higher than the energy price of the micro grid based electricity (0.201
$/kWh). In comparison with SHS, micro grid based electricity offers several benefits to the
consumers. SHS can power only limited number of appliances including few CFL/LED bulbs, radio
and a television. If users need to power many appliances then a system having a large capacity PV
system and a battery must be purchased and they are high cost, thus rarely affordable.
71
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
6 Discussion
The objective of this thesis was to explore the best renewable energy-based hybrid configuration
for powering a selected village in Burera District. That village from Burera District in the northern
province of Rwanda comprising around 150 families including several small businesses and public
utilities was selected for the analysis. The average daily load of this village is 249 kWh/day and an
approximate maximum demand of 28 kW has been observed during the evening between 18:00 –
21:00 PM. A constant load profile listed in Figure 2.4 has been assumed throughout the year because
the region is not affected by seasonal variations and the day length does not vary significantly
because Rwanda is located close to the equator.
The selected region receives an abundance of rain with an annual average of 98.6 mm. Consequently
hydropower system was used as the main resources for power production in the hybrid system, but,
due to the fact that the power need in some instances is more than what can be gained from the
available water discharge also the diesel generator and battery bank had to be used to assure at best
the supply of the peak demand. Subsequently, simulations was done for a certain number of hybrid
system configurations, and the NPV of the lifetime cost and the LCOE of each configuration have
calculated for 20 years, due to the latter parameters the lowest cost option is obtained.
Diesel powered micro grids seem to be economical based on the initial capital investment, because
they require a very small capital investment when compared to renewable energy based hybrid
systems. In contrast, hybrid systems entail a large capital investment, but lower O & M cost. Thus
the lifetime cost analysis clearly shows that the hybrid systems are more economical than diesel
powered micro grids. According to the results, the lifetime cost of a certain hybrid configuration
greatly depends on the type of generating systems involved and their rated capacities. However, the
least costly option that can meet the community’s electricity demand under the specific
requirements is the optimum solution. The decision concerning the final selection of the optimum
configuration has been made based on both optimization and sensitivity analysis, which has not
been done in most of the previous studies on hybrid system optimization using HOMER. The
optimum configuration derived here is a function of the load profile of the village and the potential
of renewable means on the site, thus this result is only valid for this site and should not be
extrapolated to other communities. According to the simulation results, the following hybrid system
configuration is found to be the optimized solution.
Micro Hydropower plant capacity
Diesel Generator capacity
Battery bank/Number of 1156 Ah batteries
Capacity of the bi-directional inverter
20 kW
10 kW
55.52 kWh/8
10 kW
The renewable fraction of the optimized hybrid system mentioned above is 0.996, hence the diesel
generator is required to supply only 843 kWh which 0.4 % of the entire load. The micro hydropower
plant generate annual energy of 198,000 kWh which is the highest percentage (99.6 %) from the
overall energy generated. This system can meet the load with an availability of around 100 %
72
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
resulting in only around 0 hours of power outage during a year, however this excludes the power
interruptions caused due to natural hazards or shut-downs for plant maintenance.
The optimal dispatch strategy of the diesel generator has been found to be “Load following” and
the generator should therefore be operated only for direct supply of the load in case of unavailability
of the renewable generation and depleted battery bank. According to the simulation results,
generator’s operating hours during one year is approximately 207 hours, this is equal to 4140 hours
in 20 years of project. The lifetime operating hours of the generator are 15000 hours more than the
operating hours of the diesel generator during the entire project. Therefore, while it is not required
to replace the diesel generator within the project life span, the battery bank should be replaced after
11 years, a procedure which costs around $ 4,430. The same case for the converter should be
replaced after 14 years, a procedure which costs around $ 4,330. The lifetime cost analysis of the
system showed that the project requires a capital investment of $ 113,000. The NPV of the lifetime
fuel cost is $ 5.300 while the total O & M cost for the whole system which primarily accounts O &
M of the diesel generator, batteries and wind turbine is $ 77,800. For the complete system, the
project is worth the NPC of $ 200,000 including the salvage values of the complex. According to
the results, the hybrid system can fulfil the demand at a LCOE of 0.201 $/kWh.
One of the important point revealed by this analysis is that, developing this power production
scheme in Rwanda requires government subsidies to make the service affordable to the customer.
The present national grid electricity price for the domestic user lies in the range of 0.2 – 2.4 $/kWh
with respect of how much kWh is consumed during one month. When compared with the latter
price, it is clear that the electricity price from the hybrid system (0.201 $/kWh) is somehow good.
However, in addition to subsidies, the active involvement of the local people to maintain the system
is very important for reducing the operating and maintenance costs. As the results show, the energy
costs can be lowered to approximately 0.13 $/kWh if the government funds become available for
covering 45 % of the capital investment and by involving the local people to maintain the system,
this can continue to decrease.
73
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
7 Conclusion
The crucial objective of this study was to find a best techno-economic of an off grid power system
to supply a rural community in Rwanda. The work was started by describing the typical load profile
of the selected community. Secondly, the identification of the possible renewable energy resources
have been done by analysing past data on the annual variations of solar and water stream in the
Rugezi catchment. Unfortunately, the wind system have not been considered because Rwanda is
closed to the equator which mean zero wind.
Off-grid renewable energy-based power systems cannot provide a continuous supply of electricity
without a storage medium. Consequently, batteries are added to the hybrid system. In order to ensure
the continuity of the supply without putting severe stress on the battery bank for a reduced overall
cost, a diesel generator are also incorporated. Further, while various component configurations for
the system have been studied, the AC coupled hybrid configuration has been selected mainly due
to its easy expandability and the maximized efficiency of the generator. After selecting the
appropriate components and studying their characteristics, the hybrid system has been modelled in
HOMER, and simulations have been made to determine the best system which can supply the
village load with the required level of availability. The usefulness cost of all hybrid structure that
can fulfil the continuous load demand has been calculated to determine the system which provides
the lowest cost.
My development of a technically feasible and economically viable hybrid solution for power
generation of one village in Burera district bring out a least-cost combination of a micro
hydropower, diesel generator and batteries that can meet the demand in a dependable manner.
A micro hydropower plant, diesel generator and a battery bank hybrid system are found as the best
option for the power system with the following capacities of 20 kW, 10 kW, and 55 kWh
respectively. The estimated value of the levelized cost of energy obtained from the lifetime cost
analysis is 0.201 $/kWh. It has been proved that the cost of energy can be further lowered
approximately to 0.13 $/kWh (90 frw/kWh) with the reduction of O & M cost and with the help of
the government donation. The energy cost of 0.13 $/kWh is acceptable and affordable for rural
consumers.
Finally, I can conclude that this hybrid power system is excellent option solution and it make a
difference to existing solutions which is more economical and attractive than grid electricity for
electrification of the selected rural community in northern Rwanda under the condition of involving
local trained people for maintaining the system and receiving some funds or donation from the
government or non-governmental organizations, with the latter condition the cost of energy for
hybrid system will be much lower than that from the national grid.
Before I propose a next steps or interesting focuses for the future work, let me first highlight a
question that arises relates to financing of the investment. Let take the case of an investment of $
113,000 will be required for a 30 kW system (or an average of $3,800/kW approximately). Even if
this volume is not big either for any normal lender (such as banks) or for any utility shareholder,
serious risks are involved in the financing. Firstly, a part of the investment is not re-deployable (e.g.
the investment for MHHP). If the project does not succeed for any reason, the investment will be
74
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
lost for the shareholder and will represent a bad investment. Secondly, the electricity market is not
good in the area and the assumptions related to the demand may not materialise, or may take longer
to accomplish. This will negatively affect the cost return mechanism. Thirdly, the business domain
may be troubled by administrative, managerial and politics objections, thereby altering such
investments. Fourthly, there are reasonable obstacles (e.g. availability of professional manpower,
govern supply logistics and bad transport means) that can attach to costs, slow up project delivery
and minimize profitability of the projects. In such circumstances, appropriate motivations and
support structure will be a paramount to attract investment and alleviate risks.
That is why the following may be an interesting focuses for the future work:
 Establishing an energy management system for the micro grid.
 Reinforcing a proper financial and business model by analyzing the economic condition in
Rwanda as well as the selected rural community in Northern Province.
 Come up with a suitable operation and maintenance scheme which can ensure the
sustainable operation of the system.
 Addressing the possibility of replacing the diesel generator in the hybrid system by locally
generated biofuels.
75
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
References
[1]
“Full
text
of
‘IPCC
report
on
SRREN.’”
[Online].
Available:
http://archive.org/stream/IPCC_REPORT/SRREN_FullReport_djvu.txt. [Accessed: 18May-2015].
[2]
D. Saheb-Koussa, M. Haddadi, and M. Belhamel, “Economic and technical study of a hybrid
system (wind-photovoltaic-diesel) for rural electrification in Algeria,” Appl. Energy, vol. 86,
no. 7–8, pp. 1024–1030, 2009.
[3]
African Development Bank, “Rwanda Energy Sector Review and Action Plan,” Report,
Tunis, Tunisia, pp. 1–108, 2013.
[4]
“Homer.” [Online]. Available: http://www.slideshare.net/WWP19910307/homer-47932711.
[Accessed: 18-May-2015].
[5]
Natinal Institute of Statistics of Rwanda, “EICV3 DISTRICT PROFILE North - Burera,”
Report, Kigali, Rwanda, pp. 1–100, 2012.
[6]
F. B. S. Mekhtoub, “PV Cell Temperature / PV Power Output Relationships Homer
Methodology Calculation,” IPCO, vol. 2, no. 1, pp. 1–12, 2014.
[7]
A. Asrari, A. Ghasemi, and M. H. Javidi, “Economic evaluation of hybrid renewable energy
systems for rural electrification in Iran - A case study,” Renew. Sustain. Energy Rev., vol. 16,
no. 5, pp. 3123–3130, 2012.
[8]
T. Dorji, T. Urmee, and P. Jennings, “Options for off-grid electrification in the Kingdom of
Bhutan,” Renew. Energy, vol. 45, pp. 51–58, 2012.
[9]
R. Sen and S. C. Bhattacharyya, “Renewable Energy-Based Mini-Grid for Rural
Electrification : Case Study of an Indian Village,” 2014, pp. 203–231.
[10] E. M. Nfah and J. M. Ngundam, “Feasibility of pico-hydro and photovoltaic hybrid power
systems for remote villages in Cameroon,” Renew. Energy, vol. 34, no. 6, pp. 1445–1450,
2009.
[11] S. M. Shaahid and I. El-Amin, “Techno-economic evaluation of off-grid hybrid photovoltaicdiesel-battery power systems for rural electrification in Saudi Arabia-A way forward for
sustainable development,” Renew. Sustain. Energy Rev., vol. 13, pp. 625–633, 2009.
[12] “NASA Surface meteorology and Solar Energy: RETScreen Data.” [Online]. Available:
https://eosweb.larc.nasa.gov/cgibin/sse/retscreen.cgi?email=rets%40nrcan.gc.ca&step=1&lat=1.30&lon=29.58&submit=Su
bmit. [Accessed: 25-Feb-2015].
[13] “Burera (District, Rwanda) - Population Statistics and Location in Maps and Charts.”
[Online]. Available: http://www.citypopulation.de/php/rwanda-admin.php?adm2id=44.
[Accessed: 17-Feb-2015].
76
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
[14] “Rugezi - Google Maps.” [Online]. Available:
https://www.google.no/maps/place/Rugezi,+Rwanda/@1.57555,29.9856891,845m/data=!3m1!1e3!4m2!3m1!1s0x19dc660fd79064ad:0xa424ce9d
3a430b1a. [Accessed: 09-Apr-2015].
[15] R. Kimera, R. Okou, and A. Ben Sebitosi, “Considerations for a sustainable hybrid mini-grid
system : A case for Wanale village , Uganda,” J. Energy South. Africa, vol. 25, no. 1, pp. 33–
43, 2014.
[16] A. Kettlewell, “The Financial Services Bill,” PFM Technical Adviser to Rwanda’s Vision
2020, pp. 1–18, 2013.
[17] L. Olatomiwa, S. Mekhilef, A. S. N. Huda, and K. Sanusi, “Techno-economic analysis of
hybrid PV-diesel-battery and PV-wind-diesel-battery power systems for mobile BTS: the
way forward for rural development,” Energy Sci. Eng., p. n/a–n/a, Apr. 2015.
[18] F. W. M. and J. N. P.M.Ndomba; D.Magoma, “Application of SWAT in Natural Wetland
Catchments. A case of Rugezi Catchment in Rwanda,” Nile Basin Water Sci. Eng., vol. 3,
no. 3, pp. 1–13, 2010.
[19] W. Resources and R. Case, “Maintenance of Hydropower Potential in Rwanda Through
Ecosystem Restoration,” 2010.
[20] K. Ajao, O. Oladosu, and O. Popoola, “Using HOMER power optimization software for cost
benefit analysis of Hybrid-solar power generation relative to utility cost in Nigeria,” Int. J.
…, vol. 7, no. April, pp. 96–102, 2011.
[21] T. Markvart and L. Castañer, Principles of Solar Cell Operation. Elsevier Ltd, 2012, pp. 7–
31.
[22] A. Mcevoy; T. Markvart and L. Castaner, Solar cells: materials, manufacture and operation.
2004, pp. 1–555.
[23] GermanSolar, “Premium Line GSM6-250P Monocrystalline Panels,” Data sheet Prem. Line,
pp. 1–2, 2011.
[24] John A. Duffie; William A. Bechman, Solar_Engineering_of_Thermal_Processes.pdf,
Second Edi. New York: John Wiley & Sons, INC., 1980, pp. 1–919.
[25] a. Drews, a. C. de Keizer, H. G. Beyer, E. Lorenz, J. Betcke, W. G. J. H. M. van Sark, W.
Heydenreich, E. Wiemken, S. Stettler, P. Toggweiler, S. Bofinger, M. Schneider, G.
Heilscher, and D. Heinemann, “Monitoring and remote failure detection of grid-connected
PV systems based on satellite observations,” Sol. Energy, vol. 81, pp. 548–564, 2007.
[26] D. Feldman, G. Barbose, T. James, S. Weaver, R. Fu, and C. Davidson, “Photovoltaic System
Pricing Trends 2014 Edition,” 2014.
77
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
[27] “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/Prod
ucts/panel018. [Accessed: 10-Apr-2015].
[28] “SMA Solar Technology AG | Sunny Tripower 15000TL/20000TL | Solar Inverter Datasheet
| ENF Inverter Directory.” [Online]. Available: http://www.enfsolar.com/pv/inverterdatasheet/501. [Accessed: 10-Apr-2015].
[29] “Solar PV ground mounted system - Google Search.” [Online]. Available:
https://www.google.no/?gfe_rd=cr&ei=VQ0oVYetBdaBNNKagbAI#q=Solar+PV+ground
+mounted+system. [Accessed: 10-Apr-2015].
[30] D. Singh, Micro Hydro Power Resource Assessment Handbook, no. September. Asian and
Pacific Center for Transfer of Technology, 2009, pp. 1–69.
[31] S. Khennas, “Micro-hydro power,” in Technical Brief, 2000, pp. 1–7.
[32] Jica, DEPARTMENT OF ENERGY ENERGY UTILIZATION MANAGEMENT BUREAU
Manuals and Guidelines for Micro-hydropower Development in Rural Electrification
Volume I, vol. I, no. June. Taguig, Metro Manila: Energy Complex, 2009, pp. 1–268.
[33] R. Kimera, R. Okou, and A. Ben Sebitosi, “Considerations for a sustainable hybrid mini-grid
system : A case for Wanale village , Uganda,” vol. 25, no. 1, pp. 33–43, 2014.
[34] Sarala P. Adhau, “Economic Analysis and Application of Small Micro/Hydro Power Plants,”
ICREPQ’09, pp. 1–7, 2009.
[35] Aaron Marks, “Reducing Diesel Consumption with AHI Batteries,” AQUION ENERGY, Las
Vegas, Nevada, pp. 1–6, 2014.
[36] “Approximate Diesel Generator Fuel Consumption Chart.” [Online]. Available:
http://www.dieselserviceandsupply.com/diesel_fuel_consumption.aspx. [Accessed: 19Mar-2015].
[37] Timothy A. Loehlein, “Maintenance is one key to diesel generator set reliability,” 2007.
[38] “Kenya Generators, Kenya Generators Manufacturers and Suppliers on Alibaba.com.”
[Online]. Available: http://www.alibaba.com/countrysearch/KE/generators.html. [Accessed:
10-Apr-2015].
[39] “Inverter, Solar Inverter - Stand-alone Inverter, Grid Tie Inverter or Grid Connected
Inverter, Hybrid Inverter.” [Online]. Available:
http://www.leonics.com/product/renewable/inverter/inverter_en.php. [Accessed: 23-Mar2015].
[40] “SMA Inverter Chargers.” [Online]. Available:
http://store.energymatters.com.au/brand/sma/products/inverter-chargers. [Accessed: 23Mar-2015].
78
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
[41] “Buy solar inverters | DC to AC power inverter information.” [Online]. Available:
http://store.energymatters.com.au/products/inverters. [Accessed: 23-Mar-2015].
[42] “Rwanda - Real interest rate (%) - Country Comparison.” [Online]. Available:
http://www.indexmundi.com/facts/indicators/FR.INR.RINR/compare?country=rw#country
=rw. [Accessed: 20-Mar-2015].
79
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Appendices
Appendix A
Survey form for Households Grid Connected
1. Form no:
2. Date:
3. Phase:
4. Number of persons in the household:
5.
6.
7.
8.
9.
Age
Number of persons Ubudehe Category (Richness classification)
0-6
1 (Poor)
7-13
2 (Low income)
14-19
3 (Medium income)
20+
4 (Wealthy)
Size of household (number of rooms):
Total household income:
Electricity costs:
When did you get connected to the grid:
Electrical appliances:
Types
Number
Usage time (hour, min)
When during the day are they used?
Lamps
Cell-Phones
Radio
TV
DVD Player
Computer
Refrigerator
Iron
Water pumps
10. Do you believe there is any difference between weekdays and weekend in your electricity
consumption?
11. How often do you get blackouts?
12. How do they affect your everyday planning?
13. Are you satisfied with the electricity distribution?
14. Do you have any plans of buying new equipment?
15. If yes, which one?
16. How would you change your consumption if the price per kWh should increase?
17. Do you use any other energy sources (firewood, paraffin, batteries)?
18. If yes, for what purpose?
 Cooking
 Entertainment (radio, TV)
 Light
 Other
19. What’s the monthly cost for these other sources?
80
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Appendix B
Detailed daily consumption for selected village
Domesticpurposes
No.
Wealthy family
Appliances
No. in use
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
6
2
1
1
1
1
1
1
1
Total Power
11
5
10
120
30
100
500
1000
500
Hrs/day
66
10
10
120
30
100
500
1000
500
5
2
12
3
3
2
4
1
1
Total
No.of houses
Total for Wealth Families
No.
Medium income family
Appliances
No. in use
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
4
2
1
1
1
0
0
0
0
Total Power
11
5
10
120
30
100
500
1000
500
Hrs/day
44
10
10
120
30
0
0
0
0
5
2
12
3
3
2
4
1
1
Total
No.of houses
Total for Medium income families
No.
Low income family
Appliances
No. in use
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
4
2
1
0
0
0
0
0
0
Total Power
11
5
10
120
30
100
500
1000
500
Hrs/day
44
10
10
0
0
0
0
0
0
5
2
15
3
3
2
4
1
1
Total
No.of houses
Total for Low income families
Watt-hrs/day TT Power Hours/day
330
660 05:00-06:00
20
100 05:00-07:00
120
100 05:00-17:00
360
1200 18:00-21:00
90
300 18:00-21:00
200
1000 17:00-19:00
2000
5000 17:00-21:00
1000
10000 09:00-10:00
500
5000 08:00-09:00
4620
10
46200
Hours/day
18:00-22:00
Watt-hrs/day TT Power Hours/day
220
1760 05:00-06:00
20
400 05:00-07:00
120
400 05:00-17:00
360
4800 18:00-21:00
90
1200 18:00-21:00
0
0
0
0
810
40
32400
Hours/day
18:00-22:00
Watt-hrs/day TT Power Hours/day
220
4400 05:00-06:00
20
1000 05:00-07:00
150
1000 05:00-20:00
0
0
0
0
0
0
390
100
39000
Hours/day
18:00-22:00
Industrial/commercial/communitypurposes
No.
Shops and Bars
Appliances
No. in use
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
4
1
1
1
1
1
1
0
0
Total Power
11
5
10
120
30
100
500
1000
500
Total
No.of houses
Total for Shops and Bars
81
44
5
10
120
30
100
500
0
0
Hrs/day
5
2
12
10
10
2
10
0.2
1
Watt-hrs/day TT Power Hours/day
220
220 18:00-22:00
10
25 12:00-14:00
120
50 10:00-22:00
1200
600 12:00-22:00
300
150 12:00-22:00
200
500 12:00-14:00
5000
2500 12:00-22:00
0
0
7050
5
35250
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
No.
Administration Post
No. in use
Appliances
Total Power
Power (W)
11
5
10
120
30
100
500
1000
200
2
2
0
0
0
2
0
0
1
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Hrs/day
12
2
12
3
3
4
4
0.2
1
22
10
0
0
0
200
0
0
200
Total
No.of Post
Total for Administration Post
No.
Medical centres
Appliances
No. in use
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
30
5
0
1
1
3
2
0
1
11
5
10
120
30
200
500
1000
500
Total Power
Hrs/day
330
25
0
120
30
600
1000
0
500
12
2
12
8
8
6
24
0.2
3
11
5
10
120
30
100
500
1000
500
Total Power
Hrs/day
220
0
20
120
30
100
0
0
500
12
1
2
2
2
4
4
1
3
11
5
10
120
30
200
500
1000
500
Total Power
Hrs/day
440
25
30
120
30
800
0
0
500
12
2
3
3
3
4
4
1
4
Total
No.of Centres
Total for Medical centres
No.
Primary Schools
Appliances
No. in use
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
20
0
2
1
1
1
0
0
1
Total
No.of Schools
Total for Primary Schools
No.
Secondary School
Appliances
No. in use
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
40
5
3
1
1
4
0
0
1
Total
No.of School
Total for Secondary Schools
82
Watt-hrs/day TT Power Hours/day
44 18:00-06:00
264
20 05:00-07:00
20
0
0
0
0
0
0
400 08:00-12:00
800
0
0
0
0
400 08:00-09:00
200
0
1284
2
2568
Watt-hrs/day TT Power Hours/day
3960
330 18:00-06:00
50
25 09:00-11:00
0
0
960
120 08:00-16:00
240
30 08:00-16:00
3600
600 08:00-12:00
24000
1000 00:00-24:00
0
0
1500
500 05:00-08:00
34310
1
34310
Hours/day
14:00-16:00
Watt-hrs/day TT Power Hours/day
2640
220 18:00-06:00
0
0
40
20 10:00-12:00
240
120 12:00-14:00
60
30 12:00-14:00
400
100 08:00-12:00
0
0
0
0
1500
500 05:00-08:00
4880
1
4880
14:00-16:00
Watt-hrs/day TT Power Hours/day
5280
440 18:00-06:00
50
25 05:00-07:00
90
30 05:00-08:00
360
120 18:00-21:00
90
30 18:00-21:00
3200
800 08:00-12:00
0
0
0
0 06:00-07:00
2000
500 18:00-22:00
11070
1
11070
14:00-16:00
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
No.
Community church
No. in use
Appliances
Power (W)
11
5
10
120
30
200
500
1000
500
12
2
1
1
1
1
0
0
1
1 Lamps
2 Cell-Phones
3 Radio
4 TV
5 DVD Player
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Hrs/day
Total Power
132
10
10
120
30
200
0
0
500
12
2
3
4
4
4
12
1
3
Total
No.of churches
Total for churches
No.
Small manufacturing units
Appliances
No. in use
1 Lamps
2 3-phases motor
3 1-phase motor
4 TV
5 Ceiling Fan
6 Computer
7 Refrigerator
8 Iron
9 Water pumps
Power (W)
3
1
1
0
1
0
0
0
0
Total Power
11
3000
1000
120
100
100
500
1000
500
Hrs/day
33
3000
1000
0
100
0
0
0
0
Total
No.of units
Total for Small manufacturing units
Appendix C
Watt
kWatt
12
4
3
3
8
2
4
1
1
Watt-hrs/day TT Power Hours/day
132 18:00-06:00
1584
10 05:00-07:00
20
10 09:00-12:00
30
120 18:00-22:00
480
30 18:00-22:00
120
200 08:00-12:00
800
0 10:00-22:00
0
0 14:00-15:00
0
500 05:00-08:00
1500
4534
1
4534
Watt-hrs/day TT Power Hours/day
396
99 18:00-06:00
12000
9000 08:00-12:00
3000
3000 13:00-16:00
0
0
800
300 08:00-16:00
0
0
0
0
16196
3
48588
Daily power hourly distribution
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
00:00-01:0001:00-02:0002:00-03:0003:00-04:0004:00-05:0005:00-06:0006:00-07:0007:00-08:0008:00-09:0009:00-10:0010:00-11:0011:00-12:012:00-13:0013:00-14:0014:00-15:0015:00-16:0016:00-17:0017:00-18:0018:00-19:0019:00-20:0020:00-21:0021:00-22:0022:00-23:0023:00-24:00
0
0
0
0
0
660
0
0
120
120
120
120
120
120
120
120
0
0
660
660
660
660
0
0
0
0
0
0
0
1760
0
0
0
0
0
0
120
120
0
0
0
0
1760
1760
1760
1760
0
0
0
0
0
0
0
4400
0
0
30
30
30
30
30
30
30
30
0
0
4400
4400
4400
4400
0
0
0
0
0
0
0
500
500
500
0
0
0
0
500
500
0
0
0
0
220
220
220
220
0
0
44
44
44
44
44
44
0
0
400
400
400
400
0
0
0
0
0
0
44
44
44
44
44
44
330
330
330
330
330
330
0
0
600
600
600
600
0
0
600
600
0
0
330
330
330
330
330
330
220
220
220
220
220
220
0
0
100
100
100
100
0
0
100
100
0
0
220
220
220
220
220
220
440
440
440
440
440
440
0
0
800
800
800
800
0
0
800
800
0
0
440
440
440
440
440
440
132
132
132
132
132
132
0
0
200
200
200
200
0
0
0
0
0
0
132
132
132
132
132
132
99
99
99
99
99
99
0
0
9000
9000
9000
9000
0
3000
3000
3000
0
0
99
99
99
99
99
99
0
0
0
0
0
100
100
0
300
300
300
300
300
300
300
300
0
0
1200
1200
1200
0
0
0
0
0
0
0
0
400
400
0
0
0
0
0
0
0
0
0
0
0
4800
4800
4800
0
0
0
0
0
0
0
0
1000
1000
0
0
0
0
0
600
600
600
600
600
600
600
600
600
600
0
0
0
0
0
0
0
500
500
500
0
0
0
0
25
25
0
0
0
0
120
120
120
0
0
0
0
0
0
0
0
20
20
0
0
0
0
0
30
30
0
0
0
0
120
120
120
120
0
0
0
0
0
0
0
500
500
500
25
25
0
0
0
0
0
0
0
0
300
300
300
0
0
0
0
0
0
0
0
25
25
0
0
0
0
0
0
0
0
0
0
0
1200
1200
1200
0
0
0
0
0
0
0
0
10
10
0
0
0
0
0
150
150
150
150
150
150
150
150
150
150
0
0
0
0
0
0
0
100
100
100
100
100
100
100
100
100
100
100
100
0
30
30
30
0
0
0
0
0
0
0
0
400
400
400
400
400
400
400
400
400
400
400
400
0
30
30
30
30
0
0
0
0
0
0
0
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
50
50
50
50
50
50
50
50
50
50
50
50
0
0
0
0
0
0
0
0
0
0
0
0
20
20
0
0
0
0
0
1000
1000
0
0
0
0
0
0
0
0
0
0
30
30
30
400
0
0
0
0
0
0
0
0
5000
5000
5000
5000
0
0
0
0
0
0
0
0
0
0
0
5000
10
10
10
2500
2500
2500
2500
2500
2500
2500
2500
2500
2500
0
0
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
500
500
500
500
0
0
2265
2265
2265
2265
2265
13670
5585
4030
19475
14085
14130
14130
6925
9925
10750
10750
5800
11300
27905
26905
25905
13255
2265
2265
2.265
2.265
2.265
2.265
2.265
13.67
5.585
4.03
19.475
14.085
14.13
14.13
6.925
9.925
10.75
10.75
5.8
11.3
27.905
26.905
25.905
13.255
2.265
2.265
83
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
Appendix D
Plots of hourly power for different elements
The energy demand for the selected village community can go beyond 30 kW due to the random
variability of the load as shown in the Figure D1 below.
40
Power (kW)
30
20
10
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure D1: The hourly power consumption in kW for AC primary load over a period of 1 year.
The following Figure D2 shows that, the micro hydropower plant will give a constant output power
of around 25 kW during a year.
35
30
Power (kW)
25
20
15
10
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Figure D2: The hourly power production of a MHPP in kW over a period of 1 year.
84
Dec
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
The following Figure D3 shows that the diesel generator will produce roughly a constant output
power of around 4 kW during a year. And let me remind that the diesel generator will be switched
on when required and the latter will produce only the required amount of power that cannot be
produced by the renewable sources or battery bank to supply the load.
8
Power (kW)
6
4
2
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure D3: The hourly power production of a diesel generator in kW over a period of 1 year.
The Figure D4 shows the output power from the converter to the grid is around 4 kW, and this
power is coming from 8 batteries in series of 1,156 Ah (6.94*8 = 55.5 kWh).
10
8
Power (kW)
6
4
2
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure D4: The hourly power production from the converter in kW over a period of 1 year.
85
University of Agder, Norway
Analysis of Power System Options for Rural Electrification in Rwanda
The Figure D5 show that the power produced by the battery bank is equal to the power used for
charging.
10
8
Power (kW)
6
4
2
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure D5: The hourly power used for battery charging in kW over a period of 1 year.
The Figure D6 here below shows that the batteries will be charged and discharged with an average
of around 85 %.
100
Battery State of Charge (%)
90
80
70
60
50
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Figure D6: The hourly battery state of charging over a period of 1 year.
86
Nov
Dec
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement