Performance Analysis of a PV Grid-connected System

at the Universidade Nacional Timor Lorosa’e

José Maria Xavier

Dissertação de Mestrado
Orientador: Prof. Álvaro Rodrigues

Mestrado Integrado em Engenharia Mecânica

Especialização em Energia Térmica

Fevereiro de 2019
Abstract

The increasing energy demand in developing countries has initiated the issue of energy
security. This has made important to develop the unexploited potential of renewable resources in
latest years, including which Solar Energy stands out. Its growing interest and enactment have led
to an increase of technological improvement of these systems, which has allowed a significant
decrease of their costs, and an increase in their economic capability. PV growth for power
generation is one of the highest in the field of renewable energies and this tendency is expected to
continue to growth in the coming years. As PV power becomes more affordable, the uses of PV
for grid-connected applications are increasing. The transition towards PV as a competitive
technology will be necessary to take place in most of the energy markets. This process requires to
be complemented by a realistic evaluation of installed power plants, with credible information and
reliable recommendations.
In this study, the evaluation of performance of a micro grid-connected system is conducted
and simulation for connected with the battery and sensitivity analysis of load fluctuation are
presented. This PV system is installed in the Faculty of Engineering, Science, and Technology
(FEST) buildings of the Universidade Nacional Timor Lorosa´e (UNTL), located in Dili, Timor-
Leste. The system is equipped with 1200 PV modules (polycrystalline silicon) using three
inverters, with a total generating capacity of 250 kW, and started its operation in late 2014.
Initially, the PV system performance from 2015 to 2016 is evaluated. Data are collected
from the FEST laboratory office combined with meteorological data. To obtain an accurate result
of the system performance, an European standard guide to assess the performance of a PV grid
connected system is used. The performance analysis result obtained through calculation of
parameters used. The simulation with battery and load fluctuation results were obtained through
HOMER Pro simulation tools. Finally, some solution to maintain a better system performance are
presented.
This study will present to the FEST of UNTL and hopefully could help in future designing,
operating and maintenance of new grid connected systems.

i
ii
Resumo

A crescente demanda de energia nos países em desenvolvimento levantou a questão da

segurança energética. Isso tornou fundamental o desenvolvimento do potencial dos recursos
renováveis inexplorados nos anos antecedentes, nos quais se destaca a energia solar. O seu
crescente interesse e implementação levaram a um aumento do aprimoramento tecnológico desses
sistemas, o que permitiu uma redução significativa de custos e um aumento da sua capacidade
económica. O crescimento da tecnologia fotovoltaica para a geração de energia é um dos mais
elevados no campo das energias renováveis e esta tendência deverá continuar a crescer nos
próximos anos. À medida que a energia fotovoltaica se torna mais acessível, os seus usos para
aplicações conectadas à rede estão aumentando. A transição para o fotovoltaico como tecnologia
competitiva será necessária na maioria dos mercados de energia. Esse processo terá que ser
complementado por uma avaliação realística do sistema instalado, com informações credíveis e
recomendações confiáveis.
Neste estudo, apresenta-se a avaliação do desempenho de um sistema micro, conectado à
rede, seguindo-se a simulação do sistema ligado a baterias com a uma análise de sensibilidade das
flutuações de carga. Este sistema fotovoltaico está instalado nos edifícios da Faculdade de
Engenharia, Ciência e Tecnologia (FECT) da Universidade Nacional Timor Lorosa´e (UNTL),
localizada em Díli, Timor-Leste. O sistema está equipado com 1200 módulos fotovoltaicos (silício
policristalino) utilizando três inversores, com capacidade total de geração de 250 kW, e iniciou as
suas operações no final de 2014.
Inicialmente, o desempenho do sistema fotovoltaico nos anos de 2015 e de 2016 foi
avaliado. Os dados foram coletados no escritório do laboratório do FECT e combinados com dados
meteorológicos. Para obter um resultado certo do desempenho do sistema, foi utilizado um guia
padrão europeu na avaliação do desempenho do sistema conectado à rede. O resultado da análise
de desempenho foi obtido através do cálculo dos parâmetros utilizados. Para a simulação com
bateria e avaliação das flutuações de carga usaram-se as ferramentas de simulação do HOMER
Pro. Consequentemente, algumas soluções para manter um melhor desempenho do sistema são
apresentadas.
Este estudo será apresentado ao FECT da UNTL e espera-se que poderá ajudar no projeto,
operação e manutenção de futuro sistemas conectados à rede.

iii
iv
Acknowledgements

First of all, I would like to thank my supervisor Prof. Dr. Álvaro Rodrigues for his
guidance, advice and support during these months of work.
I would like to thanks to the FEUP of the University of Porto for the opportunity of taking
this course.
I would like to thank the Institute of CAMOES and UNTL for the Scholarships provided
during the study program.
I would like to thanks to the Mechanical Engineering Department for the supports and
efforts provided during my study program.
I would like to thank Dr. Ruben Jeronimo for authorizing and helping on collecting data
from the FEST office.
I would like to thank Mr. Raimundo for the supporting and assistance during collecting
data in the FEST office.
I would like to thank my beloved family, Niny, Nelio, Nino and Iva and all my parents for
the inspirations and supports during my study program.
Lastly, I would like to thank my friends for the supports during the course of this project.

v
vi
Table of Contents

1. Introduction ............................................................................................................................. 1
1.1 Background ..................................................................................................................... 1
1.2 Objectives ....................................................................................................................... 3
1.2.1 General Objectives ...................................................................................................... 3
1.2.2 Specific Objectives ..................................................................................................... 3
1.3 Structure of the Dissertation ........................................................................................... 4

2. State of Art .............................................................................................................................. 5

2.1 Literature Review............................................................................................................ 5
2.2 Solar PV Technology ...................................................................................................... 8
2.3 Classification of PV Solar Cells ..................................................................................... 9
2.3.1 Crystalline Silicon (c-Si)............................................................................................. 9
2.3.2 Thin-film Technology ............................................................................................... 11
2.3.3 Organic Photovoltaic Cells ....................................................................................... 13
2.4 PV System Types .......................................................................................................... 14
2.4.1 Stand-alone PV Systems ........................................................................................... 15
2.4.2 Grid-connected PV Systems ..................................................................................... 16
2.5 Evolution of the PV Costs............................................................................................. 17

3. Energy Situation in Timor-Leste and the Potential for the Use of Renewable Energy
Resources ...................................................................................................................................... 19
3.1 Characteristics of Electrical System in Timor-Leste .................................................... 19
3.2 Renewable Energy Resources Potential........................................................................ 21
3.2.1 Solar Resources Potential ......................................................................................... 21
3.2.2 Hydro Resources Potential........................................................................................ 22
3.2.3 Wind Energy Resources Potential ............................................................................ 23
3.2.4 Biomass Energy Resources Potential........................................................................ 25

4. Methodology ......................................................................................................................... 27
4.1 The PV System Description .......................................................................................... 27
4.2 Parameters Used in the Analysis of the Performance of the PV System ...................... 29
4.2.1 Reference Yield ........................................................................................................ 29
4.2.2 Final Yield ................................................................................................................ 30
4.2.3 Performance Ratio .................................................................................................... 30
4.2.4 Total Energy Losses.................................................................................................. 31
4.2.5 Annual Capacity Factor ............................................................................................ 31
4.2.6 System Efficiency ..................................................................................................... 31
4.3 Amount of Energy Stored in the Battery when Connected to the System .................... 32
4.3.1 HOMER Pro Input Model......................................................................................... 32

vii
 4.3.2 Primary Load Input ................................................................................................... 33
4.3.3 PV Input Characteristics ........................................................................................... 34
4.3.4 Grid Input Characteristics ......................................................................................... 35
4.3.5 Inverter Characteristic Input ..................................................................................... 35
4.3.6 Solar Radiation Input ................................................................................................ 35
4.3.7 Temperature Input..................................................................................................... 36
4.3.8 Battery Input ............................................................................................................. 37
4.3.9 Control Unit .............................................................................................................. 38
4.4 Sensitivity Analysis ...................................................................................................... 38

5. Results and Discussion ......................................................................................................... 39

5.1 Solar Radiation and Ambient Temperature of the Site ................................................. 39
5.2 Electrical Load Profile for FEST of UNTL .................................................................. 40
5.3 Diurnal Average Energy Generated and Energy Consumption .................................... 41
5.4 Energy Generated and Energy Injected to the Grid by the PV System ........................ 43
5.5 Energy Supplied and Imported from the Grid to the FEST .......................................... 45
5.6 System Reference Yield, Final Yield and Performance Ratio ...................................... 46
5.7 PV System Total Losses ............................................................................................... 48
5.8 PV System Capacity Factor .......................................................................................... 49
5.9 PV Modules Efficiency ................................................................................................. 50
5.10 PV System Efficiency ................................................................................................... 51
5.11 Simulation Results of PV System with Battery ............................................................ 52
5.11.1 Simulation Result for the Average data from 2015 to 2016 ................................. 52
5.11.2 Simulation Result for 2015 data with Battery....................................................... 54
5.11.3 The Simulation Results for 2016 data with Battery .............................................. 54
5.12 The Sensitivity Analysis Result .................................................................................... 55

6. Conclusions and Recommendations ..................................................................................... 60

6.1 Conclusions ................................................................................................................... 60
6.2 Recommendations ......................................................................................................... 61

7. References ............................................................................................................................. 63

viii
List of Figures

Figure 1. Polycrystalline PV module. ............................................................................................. 9

Figure 2. Monocrystalline PV module. ......................................................................................... 10
Figure 3. An amorphous silicon PV module. ................................................................................ 11
Figure 4. Typical system of PV solar energy with battery. .......................................................... 14
Figure 5. Block diagram of an Off-grid PV systems. ................................................................... 15
Figure 6. The diagram of PV grid-connected systems. ................................................................. 16
Figure 7. PV cell prices. ................................................................................................................ 18
Figure 8. Hera Power Plant. .......................................................................................................... 19
Figure 9. Monthly solar radiation data for year 2015 to 2016 for the FEST location. ................. 21
Figure 10. Timor-Leste wind atlas. ............................................................................................... 24
Figure 11. Average wind speed for Dili, Timor-Leste.................................................................. 25
Figure 12. PV grid-connected system at FEST. ............................................................................ 28
Figure 13. PV plant mounting inclination..................................................................................... 28
Figure 14. Simulation model for HOMER Pro. ............................................................................ 33
Figure 15. Average electrical load of 2015 and 2016. .................................................................. 34
Figure 16. PV input model ............................................................................................................ 35
Figure 17. Average solar radiation values of the three years studied periods. ............................. 36
Figure 18. Average temperature of the three years period............................................................ 36
Figure 19. Total monthly energy consumption for the FEST of UNTL. ...................................... 40
Figure 20. Diurnal average energy consumption of the FEST and energy generated by the PV
system. .......................................................................................................................................... 41
Figure 21. Energy generated by the PV system. ........................................................................... 43
Figure 22. Energy injected to the grid from the PV system.......................................................... 44
Figure 23. The annual average values of Yr, Yf, and PR. ............................................................ 47
Figure 24. Monthly average PV system losses. ............................................................................ 49
Figure 25. Capacity factor of the installed PV system.................................................................. 49
Figure 26. Monthly average PV module efficiency. ..................................................................... 51
Figure 27. FEST building construction development. .................................................................. 55
Figure 28. Energy output of the PV modules. .............................................................................. 56

ix
Figure 29. PV power production and power imports from the grid.............................................. 57
Figure 30. Sensitivity analysis results of PV system plus load fluctuation but without battery. .. 57
Figure 31. The State of Charge of the UET ReFlex-100kW battery. ........................................... 58
Figure 32. The sensitivity analysis results of load fluctuation with PV system plus battery. ...... 59

x
List of Tables

Table 1. Distribution of usable biomass residues by district. ....................................................... 26

Table 2. Technical specification of PV module Kyocera KD215GH-2PB. ................................. 29
Table 3. The characteristics of the UET-ReFlex-100 kW battery. ............................................... 37
Table 4. The characteristics of the CELLCUBE FB20-100 battery. ............................................ 37
Table 5. Average monthly solar radiation and ambient temperature for FEST of UNTL. ........... 39
Table 6. Average energy supply to the FEST buildings. .............................................................. 45
Table 7. Total energy production and consumption by the FEST ................................................ 45
Table 8. Annual average of Inverter and PV system efficiency. .................................................. 52
Table 9. Summary results of simulation of PV system with batteries. ......................................... 53
Table 10. Summary simulation of 2015 and 2016 data of PV system with batteries. .................. 55

xi
List of Abbreviations

AC = Alternating Current
BPP = Betano Power Plant
CF = Capacity Factor
DC = Direct Current
EAC = AC Energy
ESC = DC Energy
Ec = Energy Consumption
Ee = Energy Export
Ei = Energy Import
EP = Energy Production
EDTL = Eletricidade de Timor-Leste
FEST = Faculty of Engineering, Science and Technology
FIT = Feed-in-tariff
Hi = Incident Energy
HPP = Hera Power Plant
IEA = International Energy Agency
INV = Inverter
LT = Total Losses
NOCT = Nominal Operating Cells Temperature
PR = Performance Ratio
PV = Photovoltaic
SOC = State of Charge
Ta = Ambient Temperature
UNDP = United Nation Development Program
VOC = Voltage Open Circuits
UNTL = Universidade Nacional Timor Lorossa'e
YA = Array Yield
YF = Final Yield
YR = Reference Yield

xii
Nomenclature

APV: Area of the PV Module (m2)

CFA: Annual Capacity Factor (%)
EAC: Energy Produced by the PV System (kWh)
Ec: Energy Consumption (kWh)
EDC: DC Energy Input of the Inverter (kWh)
Ee: Energy Export to the Grid (kWh)
Eg: Energy Generation of the PV System(kWh)
Ei: Energy Import from the Grid (kWh)
EIn: Energy Input of the Inverter (kWh)
Eout: Energy Output of the Inverter (kWh)
Ep: E nergy Produces by the PV Modules (kWh)
GO: Array Reference Irradiance (kW/m2)
HI: Incident Energy in the Array Plane (kWh/m2/day)
LT: Total Energy Losses (kWh/(kWp.d)
PO: Rated Power Output of the Installed Array (kW)
PmaxG,STC: PV Array Rated Power at Standard Condition Test (STC) (kW)
PR: Performance Ratio (%)
Ta: Ambient Temperature (0C)
YF: Final Yield (kWh/(kWp.d)
YR: Reference Yield (kWh/(kWp.d)
PV,A : Efficiency of the PV Array (%)

ηInv : Inverter Efficiency (%)

ηSys : System Efficiency (%)

xiii
xiv
1. Introduction

1.1 Background

Eelectricity became one of the most useful energy forms used worldwide. Increasing
electricity demand led to the increase of the power generating capacity. Power generating based in
fossil fuels is still dominant in the development of power plants. Even though, it can also cause
environmental problem such as CO2 emissions. Thus, it is necessary to increase the use of
renewable energy in the development of power generation system (Blaabjerg et al., 2006).
According to the International Energy Agency (2018), global electricity generating capacity has
increased by 3.1%, or about 780 TWh, in 2017. Renewable electricity shares about 25% of overall
generation. Despite solar PV and wind generation, hydro power is the largest source of
renewables-based power production with a major allocation of 65% in overall renewables output.
Large solar based power plant projects in China, Japan, USA, Germany, and India are still grown.
Many big solar projects in those countries have been commissioned, for instance, 579 MW Solar
Star project in the USA and 850 MW Longyangxing Dam Solar Park project in Cina (Chatterjee
et al., 2018).
The scenario of electricity generation strongly differs from one country to another. Power
generation in Timor-Leste is heavily based on diesel fuels. It is comprised of two main power
plants, namely Hera Power Plant (HPP) and Betano Power Plant (BPP). The capacity of HPP and
BPP are 120 MW and 130 MW, respectively. The national transmission grid is 150 kVA and has
been connected throughout the country through ten substations to power community houses.
Although the electricity is available for 24 hours across the country, the government has set its
renewable energy policy to encourage the integration of renewable energy sources such as solar
energy, wind energy and hydro power to the network, to reduce its energetic dependency and
minimize fossil fuels imports (UNDP Timor-Leste report, 2014).
Efforts in addressing renewable electricity goals are ongoing through national projects.
There were some renewable electricity projects that had been implemented in the scope of an
incentive program named Introduction of Clean Energy by Solar Electricity Generation System
project. Within this project, the first solar power plant with the capacity of 250 kW was built in
2014. This solar PV grid-connected system is located in the Faculty of Engineering, Science, and

1
Technology (FEST) of the Universidade Nacional Timor Lorosa’e (UNTL) campus. In addition to
this PV grid-connected system, small size (5 x 7 kW) PV Stand-alone systems were also built in
five public schools in different locations (Timor-Leste Government, Ministry of Education. 2014).
The introduction of these advanced renewable energy technologies provides those institutions with
energy and reduces the negative impacts of using fossil fuels to generate electricity. In addition to
the reduction on fossil fuels consumption, it offers those institutions knowledge on sustainable
energy options and demonstrates how these technologies can be implemented in the local context.
Moreover, it provides those institutions with a clear evidence on how far these technologies can
address their electricity problems.
Beyond solar projects, a mini hydropower project was also built. The Norwegian
Government, through its cooperation program with the Timor-Leste Government, had installed a
micro hydroelectric with the capacity of 326 kW. This system was integrated to the main grid to
power communities’ households in the Suco of Gariuai, Baucau District. The Gariuai Mini
Hydroelectric Power project is the first mini hydropower station in the country. Electricity
produced by this system is counted about 1.5 GWh annually (Hoeiseth and Klassius, 2007). The
implementation of this project has positive impacts on local communities. It improves the quality
of life in individual household. Local people were directly involved in the construction of the micro
hydropower station. Beyond the job opportunities, the application of the hydropower station had
responded to the local electricity demands. However, unfortunately, the system is now
disconnected from the main grid because it has been broken. This occurred because of a natural
disaster happened in 2012. The authority faces difficult in fixing the system due to lack of human
resources and financial support, such as cost for replacing some damaged parts. Local communities
are now having access to the electricity from the main grid because the government has expanded
power distribution line to this community (Timor-Leste Government, Office of EDTL. 2018).
Increasing the knowledge about the technologies used to convert renewable sources of
energy into electricity is crucial for the development of the sector in Timor-Leste and can
contribute positively to economic growth and welfare of the population.

2
1.2 Objectives

It is undeniable that solar energy available at a specific site is not same throughout the year.
It varies also from one location to the other. Apart from this, seasonal variation also affects the
amount of solar radiation reaching the surface of the earth. This solar radiation would directly
affect the amount of energy generated from the power plants. Likewise, power production from
solar PV system depends on the quality of components used in the system. In order to maintain
continuous operation of a solar PV plant, it is important to carry out an effective study on the
system operating characteristics under the weather conditions, and components used in the
installed system. Only in this way the performance of the installations can be improved and the
penetration of the technologies can be increased.

1.2.1 General Objectives

The general objectives of this dissertation are to obtain data on the generation of electric
energy from the solar PV plant installed on campus of FEST-UNTL and perform an analysis to
see whether the system installed is performing satisfactorily. Furthermore, storage, by means of
batteries, is a potential improvement that will also be carried out. The motivation for this exercise
is the fact that the energy not consumed in the faculty buildings during day time is fed into the grid
but not paid, whereas the energy consumed during night time is paid by the faculty. In addition
to this, an analysis based on the data obtained will also be carried out to see the impact of the
available system capacity with the increasing demand in the future.

1.2.2 Specific Objectives

The learning objectives that specifically can be achieved from this dissertation process are
to:
Qualitatively evaluate the performance of the system by using the performance standard guide
developed by the International Energy Agency (IEA);
Quantify energy produced, through monitoring data collection system, from the system installed in
the FEST of UNTL at Hera Campus;
Quantify energy consumed of the FEST from the system;
Provide a clear information on how this system can help FEST in reducing their energy bills;
Discuss the need to expand the system, installing more generating capacity;

3
 Ascertain the prospect for using batteries in responds to the future increase of FEST energy
demand;
Identify other renewable sources, such as wind energy, to complement the system.

1.3 Structure of the Dissertation

This dissertation is structured as follows:

Chapter 1 presents an introduction, which explains the background of the power generating
in Timor-Leste, the objectives of this study, and the specific objectives that can achieved.
Chapter 2 provides a state of the art, which presents a literature review on PV grid-
connected system performance analysis. Some examples of PV grid-connected system
performance have been investigated and are discussed. The PV technologies, system types, and
evolution of the PV costs are also assessed.
Chapter 3 presents the energy situation in Timor-Leste and the potential for the use of
renewable energy resources. Electrical system characteristics and the potential of renewable
energy sources available in the country are explained. The environmental impact from those
resources will be briefly discussed.
Chapter 4 presents the methods that are used to answer to the specific questions described
in this dissertation. The system description and the parameters that are used to analyze the
performance of the Hera PV plant are explained. The input parameters for the simulation of the
system with battery, and the sensitivity analysis of load fluctuation are presented.
Chapter 5 presents the results and discussion. Energy generated and supplied by the system
is discussed. The performance analysis results based on the parameters provided are explained and
discussed, and the simulation of batteries attached to the system, and the sensitivity analysis of
load fluctuation are explained.
Chapter 6 sets out the final conclusions of all the work, summarizes the findings and the
answers to key questions in this dissertation.

4
2. State of Art

2.1 Literature Review

Literature review on the previous works related to the performance analysis of a solar PV
grid-connected systems is done in this section.
Perez et al. (2007) presented the performance analysis of the 200 kWp PV grid-connected
installation installed in Jaén University Campus, Spain. The PV installation consists of four
systems with different architecture and configuration. System 1 and 2 were installed in the
university parking area, system 3 was installed in the university buildings where the inverters, the
data monitoring system and the safety and protection system were also located. Meanwhile, the
PV system 4 was integrated in the south facade of an existing building close to the system 3. All
systems were completely integrated at campus buildings. This study showed the system has
worked effectively. Also, the authors found that PV systems 1 and 3 have shown good performance
whereas systems 2 and 4 indicate a reduction of performance ratio. In order to increase the
performance ratio of the system, the defective part of the installation which was responsible for
the diminution of PR was replaced.
Karki et al. (2012) conducted a comparative study of performance of PV grid-connected
system in two different cities, Kathmandu and Berlin. Authors analyzed the performance using
PVSYST software. A 60 kWp system was simulated with the same parameters in both cities. The
amount of energy produced from PV array and the energy fed into the grid were also investigated.
Apart from the energy produced, various energy losses through the system were also analyzed.
Some studies were carried out by Kandasamy et al. (2013) on a 1 MWp grid connected
solar power plant at four different cities in India. Authors conducted a detailed study on the solar
potential assessment for those sites. PV SYST software was used to analyze the system
performance. A comprehensive evaluation of energy produced, efficiency of PV array, system
performance ratio and other cost parameters were executed.
Sharma and Chandel (2013) conducted a detailed study on a PV grid interactive system
installed at Khatkar-Kalan, India. In this study, authors analyzed the performance of a 190 kWp
PV power plant by using two different methods, PV SYST software and practical measurements.
The PV system capacity factor, the average performance ratio, and the system efficiency were

5
elaborated. In addition, total estimated system losses due to temperature, solar radiation, array
mismatch, module quality, inverter, and ohmic wiring were also investigated.
Kazem et al. (2014) designed, implemented, and evaluated a 1.4 kW grid-connected PV
system in Sohar, Oman. Two parameters were used to evaluate the system, the capacity factor and
the yield factor. The cost of energy as well as simple payback period were also calculated. The
result shown the capacity factor and the yield factor for the proposed system were 21% and
1875kWh/kWp/year, respectively. In terms of the economical evaluation factor, it was found that
the cost of generating energy from the PV system was lower compared to the subsidized price of
energy generated with fossil fuel. The cost of energy generated by the system was about 0.045
USD/kWh with the simple payback period of about 11 years.
Kumar et al. (2014) carried out a study on a roof top 20 kWp PV grid connected system in
a manufacturing industry in India. In this study, the operational behavior, economic calculation,
and some significant features of the installation were emphasized. Moreover, important aspects of
PV plant installation such as the viability of locality in terms of geographical data, solar panel
inclination design and interfacing aspects of PV system with network were also discussed. The
detail of the outcomes obtained from the study such as monthly energy generation, maintenance
aspects, performance ratio, capacity factor, economic scrutiny of the system were also elaborated.
Kumar and Shudakar (2015) carried out a performance study of a 10 MWp grid connected
solar photovoltaic power plant installed at Ramagundam, India. In this study, solar PV plant design
aspects along with its annual performance were developed. Various system losses due to
temperature, internal network, inverter and ohmic wiring were examined. Apart from these losses,
the system performance ratio and the energy final yield were also calculated. The performance
results were then compared with the simulated values attained from the PV SYST and PV-GIS
software. The results have shown the final yield of the plant varying from 1.96 to 1.07 h/d, and
annual performance ratio of 86.12%. The annual amount of energy generated from the plant was
about 15.7982 GWh.
Similar study was carried out by Attari et al. (2016) on a 5 kWp PV grid-connected system
installed in Tangier of Morocco. The system was installed in top of a government building and
made up of 20 modules of 250 Wp. The experimental data during the periods of 2015 was recorded
based on the real time surveillance. In this study, energy output of the system, final yield, modules
temperature, plant efficiency, and system performance ratio were assessed. Various power losses

6
were also investigated. The result of the study have shown, the final yield range between 1.96 to
6.42 kWh/kWp, annual capacity factor is 14.83%, plant efficiency is 11.99%, and the performance
ratio varying from 58% to 98% depending on the irradiation levels. In addition, the plant supplied
energy to the grid with 6.4113 MWh during the year 2015.
Shukla et al. (2016) showed the simulated performance of a 110 kWp solar rooftop
photovoltaic plant connected to distribution network for a residential hostel building at Manit,
India. The authors analyzed the viability of solar PV system at the given location. In this study,
four types of modules were used in the simulation to examine system performance ratio and energy
yield. In addition, the Solar GIS PV Planner software was used to evaluate the performance of the
system. Global in-plane radiation, shading effects, angular reflection at the surface of the arrays,
system losses, and power conditioner performance were evaluated. The study result have shown
that, the performance ratio ranged between 70% to 88%, and the energy yield ranged from 2.67
kWh/kWp to 3.36 kWh/kWp. Furthermore, among the four types of modules examined, a-Si and
CdTe PV system have performance ratios greater than 75%.
Allouhi et al. (2016) investigated the performance analysis, economic and environmental
evaluation of two 2 kWp solar power plant installed in High School of Technology of Meknes,
Morocco. In this study, poly-Si module and mono-Si module were used, and the simulations were
carried out using PVSYST software. The meteorological data for the site were taken from
METEONORM database. System capacity factor, final yield, and system efficiency for both
installed technologies were calculated. From the simulated results, the authors found that PV
installation using Poly-Si technology has a higher performance compared to the Mono-Si
technology. The average daily final yields for Poly-Si and Mono-Si were, respectively, 4.98 h/year
and 4.85 h/year. Furthermore, the annual overall efficiency and annual capacity factor of the PV
system were 20.52% and 12.3% respectively.
Kumar et al. (2017) analyzed a simulated performance on a 100 kWp Si-poly PV grid-
connected system. The authors simulated the proposed system to evaluate the feasibility of
installing that system in an educational institute. The simulated system consists of 323 Si-poly
modules of 310 Wp. All modules were arranged in 17 strings and each string was made up of 19
modules connected in series. Four solar inverters of 20 kW were used to connect the system to the
main network through a utility meter. The PV SYST V6.52 software was used in the simulation.
The effective energy output, energy supplied to the grid, and performance ratio of the system were

7
evaluated. In addition, the system losses were also processed. The simulation result shown, the
system produces 165.38 MWh/year, and 161.6 MWh/year was injected to the grid. Moreover, the
annual performance ratio was about 80%.

2.2 Solar PV Technology

Among the clean energy resources, the conversion of the solar radiation into electricity by
PV devices is a reliable choice to meet the global energy demand. Solar PV technology is one
among other technologies to convert renewable resources which have a potential to offer clean
energy, affordable, and reliable electricity system in the future (Tyag et al., 2013). Despite the
benefits of this clean and unlimited resource, a higher efficiency to cost ratio is necessary for the
installed PV devices to make them competitive with the conventional energy resources. Observing
this fact, governments all over the world are inspired to develop and deploy solar PV technology.
Sampaio and Gonzales (2017) reported that, today, there is a wide variety of PV cells technology
available in the market that utilize different type of materials and it will increase in the future. PV
cell technologies are normally classified in three generations, based on the raw material used and
on the level of commercial maturity. The three generations are as follows:

First generation PV systems, fully commercial, use the wafer-base crystalline silicon
(c-Si) technology, both in its simple crystalline form (sc-Si) as well as in the multi-
crystalline form (mc-Si).
Second generation PV systems, in their early market deployment phase, are based on thin
film photovoltaic technologies and generally include three main families: (1) Amorphous
silicon (a-Si) and micro amorphous silicon (a-Si /μc- Si); (2) cadmium telluride (CdTe);
and (3) copper indium selenide (CIS) and copper, indium gallium dieseline (CIGS).
Third generation PV systems involve organic photovoltaics cells and concentrating PV
(CPV) cell technologies that are still in demonstration or have not been widely marketed
and new concepts in development.

Goetzberger et al. (2002) reported that there are some requirements for a solar cell material
to be considered ideal: (1) band gap between 1.1 and 1.7 eV, because the smaller the gap, the easier
it is to promote an electron from one band to the other and thereby increase the conduction of this

8
material; (2) direct band structure; (3) consisting of readily available materials, non-toxic; (4) easy
fabrication technique, suitable for large production volumes; (5) good photovoltaic conversion
efficiency; (6) long-term stability.

2.3 Classification of PV Solar Cells

2.3.1 Crystalline Silicon (c-Si)

Silicon is one of the leading materials in solar PV technology. The first-generation PV

modules was manufactured from crystalline structure of silicon. Crystalline silicon has been the
major material for the manufacture of solar cells for the past 20 years and it is expected to continue.
This is based on the inherent benefits of silicon as a semiconductor and on the major industry that
has been built to manufacture silicon devices (Bruton, 2002). The c-Si has two types that are
available in the market, polycrystalline and monocrystalline.

Polycrystalline
Polycrystalline silicon usage is growing more rapidly than monocrystalline. This is because
polycrystalline silicon has lower capital cost for wafer production, higher silicon utilization and
square wafer, which give a higher packing density in the module. Polycrystalline has lower cell
efficiency. A study conducted by Zhao et al. (1998) reported that the honeycomb textured
polycrystalline solar cell efficiency was 19.8%. A polycrystalline silicon module can be viewed in
figure 1.

Figure 1. Polycrystalline PV module.

9
Monocrystalline
In contrast, monocrystalline wafer is expensive to produce but gives a higher efficient solar
cell. A study conducted by Zhao et al. (1998) reported that the honeycomb textured
monocrystalline solar cell efficiency was 24.4%. However, its module efficiency was lower. A
monocrystalline silicon module can be viewed in figure 2.

Figure 2. Monocrystalline PV module.

Although the percentage of polycrystalline silicon solar cells production is likely to rise, a
market for high-efficiency monocrystalline silicon solar cells will always exist. High efficiency is
important in applications where space is limited, higher labor cost, and the amount of solar
radiation is relatively low (Bruton, 2002). Zhao et al. (2001) reported that, passivated emitter, rear
totally diffused (PERT) and passivated emitter, rear locally diffused (PERL) cells with a high-
efficiency have been fabricated on magnetically confined Czochralski (MCZ) silicon wafers. A
24.5% energy conversion efficiency based on PERT MCZ had demonstrated and it was confirmed
by the Sandia National Laboratories under the standard test condition of 100mW/cm2 global
AM1.5 spectrum at 25oC.
Crystalline silicon cells in mass production have considerably lower efficiency. For
polycrystalline with large grains, the efficiency is in the range of 15 to 17 % and for
monocrystalline one in the range of 17 to 19 %. However, the cost per Watt peak of the PV system
is too high to compete with the traditional energy sources, fossil fuels. Therefore, the efficiency of
cells should increase, and the areal cost of the PV system should decrease. The silicon solar cells

10
produced today have a potential to increase the efficiency and to reduce the price, but there is a
limit of the cost which could be too high. To overcome this limit, the new concepts of solar cells
of third generation photovoltaics are developed (Green, 2006).

2.3.2 Thin-film Technology

Thin-film is an alternative technology, which uses less or no silicon in the manufacturing

process. Today, an extensive variety of TF-Si PV devices have been developed on the commercial
scale, including single-junction amorphous silicon (a-Si), dual junction a-Si/a-Si, tandem-junction
microcrystalline silicon, amorphous silicon known as “micro-morph”, and triple-junction
germanium-doped amorphous silicon (a-Si/a-SiGe/a-SiGe) (Gul et al., 2016). Some types of thing
film technology are described in this sub section.

Amorphous Silicon (a-Si)

Amorphous silicon is a non-crystalline allotropic form of silicon and the most commonly
developed. It is well known among thin film technology, but it is prone to degradation. Some of
the varieties of a-Si are amorphous silicon carbide (a-SiC), amorphous silicon germanium (a-
SiGe), microcrystalline silicon ( -Si) and amorphous silicon-nitride (a-SiN) (Parida et al., 2011).
The first a-Si publications related to the manufacture of solar cells appeared after the 1960s, and
it was initially reported by Carlson in 1976. An a-Si occurred in the market in 1981 (Sampaio and
Gonzales, 2017). Amorphous silicon module can be seen in figure 3.

Figure 3. An amorphous silicon PV module.

11
 An a-Si cell is used for single-junction cells but tend to give efficiencies of only about 6%
or less because of high defect concentrations related with the lack of crystallinity (Lechner and
Schade, 2002). Research was continuing to find ways to reduce this effect during the last two
decades. The highest confirmed stable efficiency for an a-Si cell was about 9.8% (Green et al.,
2011). The high expectation of this material was contained by the relatively low efficiency
obtained so far and by the initial degradation induced by light. This is the reason why a-Si has not
been able to reach a significant share in the PV global markets.

Cadmium Telluride (CdTe) or Cadmium Sulphide (CdS)

Cadmium telluride solar cells are formed from cadmium and tellurium. The CdTe has been
known to have the ideal band gap of 1.45 eV with a high coefficient of absorption of the solar
spectrum being one of the most promising photovoltaic materials for thin film cells. However, the
toxicity of cadmium (Cd) and environmental issues related to the use of this material pose a
problem for this technology. Therefore, First Solar, one of the world's largest manufacturers of
photovoltaic solar modules, has launched a recycling program for deactivated PV cells, extremely
popular in the field of thin films because of the efficiency of its process, which has the capacity to
reduce the cost of production to make this technology more competitive. The other potential
problem is the availability of Te, which can lead to scarcity of raw materials, thus affecting the
cost of the modules (Sampaio and Gonzales, 2017). There were several remarkable studies
conducted previously on CdTe cells. Amin et al. (2001) found the CdTe efficiency of 10.6% and
11.2% was obtained on a thin film of 0.55 and 1 mm thick, respectively. In addition, CdTe on
plastic foil with an efficiency of 11.4% is reported by Upadhayaya et al. (2007). In general, 15%
to 16% cell efficiency has been obtained by Ferekides and Britt (1994), and Wu et al. (2001). In
July 2011, First Solar company sets the world record of 17.3% cell efficiency, which was
confirmed by the National Renewable Energy Laboratory, NREL (National Renewable Energy
Laboratory, 2015).

Copper Indium Gallium Selenide (CIGS) or Copper Indium Selenide (CIS)

Copper and indium diselenide (CuInSe2) or indium copper selenide (CIS), as is sometimes
known, and copper-indium-gallium selenide (CIGS) are photovoltaic devices containing
semiconductor elements of groups I, III and VI of the periodic table which are Beneficial because

12
of their high optical absorption coefficients and their electrical characteristics that allow the
adjustment of the device (Hosenuzzaman et al., 2015). Some of the major challenges of these
technologies have limited ability to expand the process of high yield and low cost, degradation
under wet conditions, as it promotes changes in the properties of the material and the shortage of
Indian in nature (Chaar et al., 2011). Repins et al (2008) reported that, the CIGS cell efficiency in
2006 was 20% and about 13% for module (Powalla, 2006). In the late of 2013, Siva Power has
reported highest cell efficiency of 18.8% which was confirmed by the NREL (Osborne, 2014).

Gallium Arsenide
Gallium arsenide (GaAs) is a compound semiconductor form of Gallium (Ga) and Arsenide
(As). GaAs has a similar structure as silicon cells with high efficiency and less thickness, and it is
lighter as compared to monocrystalline and multi-crystalline silicon cells (Iles, 2001). Its energy
band gap is 1.43 eV (Streetman and Banerjee, 2005) which can be improved by alloying it with
Aluminium (Al), Antimony (Sb), Lead (Pb), which in turn will form a multi-junction device
(Satyen, 1998). The Spire Corporation has manufactured the most efficient triple-junction, GaAs
cell, with an efficiency of 42.3% (Osborne, 2010). The Sharp company has been pursuing research
and development of a triple-junction compound solar cell that achieves high conversion efficiency
by stacking three photo-absorption layers. In 2009, Sharp succeeded in improving cell conversion
efficiency to 35.8% based on proprietary technology that enabled efficient fabrication of a stacked
triple-layer structure with indium gallium arsenide (InGaAs) as the bottom layer. In 2011,
conversion efficiency of 36.9% was achieved, at the research level, for a triple-junction compound
solar cell (Sharp, 2011). This technology is still under research and, hence, there are fewer
commercial modules available in market.

2.3.3 Organic Photovoltaic Cells

Organic photovoltaic cells offer the long-term potential of achieving the goal of a PV
technology that is economically viable for large-scale power generation, since organic
semiconductors are a less expensive alternative to inorganic semiconductors, such as silicon.
Furthermore, organic molecules can be processed by simpler techniques that are not suitable for
crystalline inorganic semiconductors. Almost all organic solar cells have a flat layered structure,
wherein the light absorbing layer is sandwiched between two different electrodes. One of the

13
electrodes has to be semi-transparent, the indium tin oxide (ITO) is normally used, however a thin
layer of metal can also be used. Calcium, magnesium, gold and aluminum can also be used as
electrodes, the latter being the most used (Benanti and Venkataramen, 2006).
Research on organic solar cells aims to increase the conversion efficiency of solar energy,
since the total energy output of a solar cell is equal to the product of its efficiency and lifetime.
Therefore, the stability which is directly related to the lifetime is an important property for this
type of cell, since it impacts the value of an energy production system based on this technology.
Over the past few years, many aspects of organic solar cells have been extensively studied,
including the synthesis and application of new materials, physical process modeling, large-scale
manufacturing, and improved stability (Cao et al., 2014). However, the research and development
of organic solar cells still has a long way to go to compete with inorganic solar cells (Benanti and
Venkataramen, 2006).

2.4 PV System Types

Solar PV system can be designed across a wide range of complexity, from a very simple
system which simply designed to feed a load, to a system with a high degree of complexity such
as large PV plants. A typical stand-alone PV system consists of four basic elements; PV modules,
charge controller, the inverter, and the battery when required. One of these systems is sketched in
figure 4.

Figure 4. Typical system of PV solar energy with battery.

14
 The PV module consists of PV cells, which convert directly solar energy into DC
electricity. These surfaces have no moving parts to wear out and work without the use of fuel,
without vibrations, no noise and not harming the environment (Lan and Li, 2014). As for the charge
controller, it has the function to preserve the batteries from being overcharged completely and
increasing battery useful life. The inverter, in turn, is responsible for converting the power
generated by the PV modules into alternate current (AC). Batteries are used in a PV system to
store the excess energy generated by the modules to be utilized at night or on days with low
sunshine (Hosenuzzaman et al., 2015). There are more complex of PV systems such as stand-alone
systems, on-grid connected systems and hybrid systems.

2.4.1 Stand-alone PV Systems

Stand-alone PV systems are not connected to the local grid. They are beneficial in remote
areas that are isolated from the power distribution network, and are mainly used for small self-
consumption. For remote areas where the AC main grid is not accessible, the stand-alone PV
system provides energy to local users behaving as an AC voltage source. Because of the
unpredicted nature of the PV source, a chargeable battery or backup supply is necessary in order
to store the excess energy during the high solar irradiation period and supply to the load when the
PV energy is not available. Fluctuating voltage and power are of major concern in the PV stand-
alone systems. To address these issues, a control strategy for voltage control using voltage source
inverter in the voltage control mode is used. The block diagram of a common stand-alone PV
system is illustrated in figure 5.

Figure 5. Block diagram of an Off-grid PV systems.

15
 The standard voltage level of such operations is normally in the range of 220-250 AC.
Consequently, the voltage level of the DC link of the DC-AC inverter that feeds these loads needs
to be maintained around 360-400 V. The voltage levels of the PV modules that are available
commercially are generally in the range of 12-15 V. Therefore, to form a DC bus of around 360-
400 V, several PV modules are to be connected in series to form PV string (Walker and Sernia,
2004). Variable DC link voltage across the inverter can affect the sensitive load supplied. As there
are no grid lines in an off-grid PV system, the output voltage has to be controlled in terms of
amplitude and frequency. The off-grid PV system control is featured with a frequency and output
voltage controller that is capable of handling variable loads.

2.4.2 Grid-connected PV Systems

Grid-connected PV systems are directly connected to the distribution grid for the supply of
surplus energy into the grid after self-consumption. These types of PV plants are useful for high
power generation and efficient use of solar power. In addition to that, it also uses in the residential
or institution with the objectives of reducing their electricity bills and promoting clean energy. In
grid-connected PV systems batteries are not needed, since all power generated by the PV plant is
uploaded to the grid for direct transmission, distribution and consumption. Hence, the generated
PV power reduces the use of other energy sources feeding the grid such as fossil fuels. Since grid-
connected system do not need batteries, they are more cost effective and require less maintenance
and reinvestment as off-grid connected PV system do. The grid generally refers to the power
distribution system, which receives its input power from substation at 440V and 220-250V single-
phase AC, at 50 Hz. Individual consumers can utilize the power in the range of 10-15 kW (Oriti
and Julian, 2011). The diagram of a PV grid-connected system is shown in figure 6.

Figure 6. The diagram of PV grid-connected systems.

16
 The main components in a PV grid-connected system are the inverter and the modules. The
inverter converts the available DC power from the PV array into the usable AC power consistent
with voltage and power quality requirements of the grid utility. A bidirectional interface is made
between the PV system AC output terminals and the grid utility network. This enables the PV
system to supply power to the local loads of feed to the grid, when PV power is greater than load
demand power. At night, or during high load demands, the power required by load is greater than
the PV power generated. The excess power required by the load is received from the grid utility.
This safety feature is required in all grid-connected PV systems, and ensures that the PV system
will not continue to operate and feed into the grid when the grid is down for service or repair (Oriti
and Julian, 2011).

2.5 Evolution of the PV Costs

Costs of manufacturing of PV module and the price of the PV module have decreased
significantly for decades. The first considerable reduction in PV module costs happened in the
middle of 1970s, when PV moved from space to terrestrial applications. This revolution allowed
for reduced demand for device quality and reliability, greater standardization and increased market
competition (Maycock, 2011). This reduction in costs was caused by a decrease in c-Si module
prices from $90 per watt-peak in 1968 to a $15 per watt-peak in 1978. The decline of c-Si costs
continued over the periods, and improved device efficiency and manufacturing scale were judged
to be the main cost reduction drivers, accounting for 30% and 40% of reduction respectively
(Nemet, 2006).
Mints (2009) reported that the PV price continued to fall and stood at about $9 per watt-
peak in 1987 and then increased from 1988 through 1990 as the PV modules were reduced due to
a limitation in the availability of silicon wafers. PV module prices was then decreased substantially
from 1991 up to 1995 because of increases in PV module manufacturing capacity and a worldwide
recession that slowed PV demand. PV module prices continued to fall slightly from 1995 to 2003,
which was due to increases in module capacity globally and a growing PV market. PV module
prices were less than $5 per watt-peak by 2000, and even reaching a low of $3 per watt-peak in
2003.

17
 Donne et al. (2013) reported that module prices in 2012 for the Unites State of America
(USA) and Europe per watt peak were about $2.29 and €2.17 respectively. The PV cell prices trend
from 1980 to 2014 can be seen in figure 7.

Figure 7. PV cell prices.

The cost of a complete PV system including additional equipment such as PV modules,

inverter, charge controller, electrical components and mounting structure were also decreased.
Factors such as increases in the amount of production, improvement in the efficiency by innovative
technology, innovation in material technology, increasing lifespan of PV system, and policies for
solar technology contributed to the decline of the PV costs (Green Peace International, 2011).
Gul et al. (2016) reported that the PV system costs vary from country to country and it also
depends on the size of the installed systems. For example, China was the cheapest country, in terms
of complete installation of PV project, and South Africa was the costliest, accounted for about
£0.32 per Watt and £4.29 per Watt respectively. Considering the size of project, for 5 MW PV
plants installed in the Panama was the costliest as compared to China, and India. For 50 MW
projects, Chili was the most expensive as compared to India, Equador, and South Africa.

18
3. Energy Situation in Timor-Leste and the Potential for the Use of
Renewable Energy Resources

3.1 Characteristics of Electrical System in Timor-Leste

Electricity generation in Timor-Leste is powered by engine diesel fuels. Timor-Leste

imports all of its fossil fuels including diesel, gasoline, liquid petroleum gas (LPG) and kerosene
from overseas. Electricity utility is owned by the state. Presently, Timor-Leste has two power
generating stations, the Hera Power Plant (HPP) and Betano Power Plant (BPP). The HPP is
located in Dili, the capital of the country, while BPP is located in Betano. Betano is located in the
north part of the Dili District. The Timor-Leste state owns these plants. The HPP can be viewed in
figure 8.

Figure 8. Hera Power Plant.

HPP has a total capacity of 119 MW (7 x 17 MW), and BPP has a total capacity of 136
MW (8 x 17 MW). Both have one sub-station that rises the voltage to 150 kV, for the purpose of
connection with the transmission system. Nine substations to reduce the voltage also being built
in district capitals of Timor-Leste. These substations allow connection to the existing lines of 20
kV distribution. The power distribution system carries power from the substations to supply
consumers.

19
 The management of the utility and services to the consumers is responsible by the
Electricidade de Timor-Leste (EDTL). EDTL is a government department under the Ministry of
Public Works, Telecommunication, and Transportation. Wartsilla in a consortium with Puri
Akraya Engineering Limited are now responsible for operating the power plant, and for all service
and maintenance activities that ensure plant efficiency and availability. Puri Akraya Engineering
Limited is an Indonesian company, Wartsilla is a Finish company and the equipment manufacturer.
The operation and maintenance of national transmission system is performed by the China Nuclear
Industry 22nd (CNI22). CNI22 is a Chinese company who built the transmission lines. Diesel fuels
for reliable and stable operation of the power plant have been supplied by the Esperança Timor
Oan Limited (ETO). ETO is a Timorese company.
Effort has been made by the government to supply power to rural consumers all around the
country. At present day, some parts of rural areas still do not have access to electricity from the
main grid. Households settlement and the topography of the site are two mains obstacles to extend
the distribution line. The majority of households across the country have now access to the
electricity for 24 hours per day. This achievement has carried a considerable and potentially
increasing financial cost to the government. A substantial proportion of this cost is from the cost
of the subsidization of the provision of diesel-generated electricity. This is mainly driven by the
high fuel cost per unit of electricity generated and the low rate of cost recovery.
Regarding electricity generated based on renewable energy sources, the Timor-Leste
government has the commitment to introduce clean energy through the deployment of the
renewable energy technologies. Over the past four years, government has constructed a few small
PV stand-alone systems and one PV grid-connected system. Those systems were installed in the
public institutional education buildings at different locations. Hera PV grid-connected system was
one of the projects. This project was built early 2014 and started its operation in November 2014.
Although, the PV grid-connected system was implemented but the country does not have the feed-
in-tariff (FIT) policy. Consequently, the Hera PV system owner, in this case FEST, does not get
any money for the amount of power the system supplies to the electrical system. Unluckily, FEST
must pay for the amount of electricity that imports from the grid.

20
3.2 Renewable Energy Resources Potential

Timor-Leste is a country with a great potential in the use of renewable energy resources.
These renewable resources should be exploited to reduce its dependency on fossil fuels. Despite
fossil fuel cost dependency reduction, it contributes to environmental and economic benefits.
Renewable resources available in the country are discussed in this section.

3.2.1 Solar Resources Potential

Solar energy can be defined as the description given to any form of light energy and thermal
energy obtain from the sun, and subsequent transformation of that energy into some form that can
be used, either directly for water heating or such as electrical or mechanical energy (Martifer,
2010).
Solar energy potential is the highest among the various renewable energy resources in the
country. Timor-Leste lies near to the equator line where length of daylight is almost constant for
most of the year. Average monthly solar radiation per square meter from 2015 to 2016 for the
FEST is presented in figure 9.

Figure 9. Monthly solar radiation data for year 2015 to 2016 for the FEST location.

21
 As it is observed from figure 9, solar radiation reaching the plant is different in 2015 and
2016. The highest solar radiation occurred in the months of August and September, accounted for
about 195.50 and 187.21 kWh/m2, respectively. The monthly solar radiation for the other months
recorded more than 120 kWh/m2. Annual total radiation for the two periods was about 1.94 and
1.97 MWh/m2, respectively. While the average total radiation was about 1.95 MWh/m2.

Environmental Impact
Solar power plant projects are not emissions free at all. Emissions associated with solar PV
during the manufacturing are mercury emissions, cadmium, and Green House Gas (GHG)
emission. Emissions of mercury and cadmium from a solar PV system are about 0.1 g Hg/GWh
and 0.02 g Cd/GWh respectively. Every 100 MW of solar power plant emits 15.6 gCO2/kWhe
(Tripathi et al., 2016). Furthermore, solar power plants require an amount of land area. The
dimension of the land area depends on the capacity of the PV plants and the specific PV technology
used. For a small size of PV system capacity, typically can be placed on the top of an existing
building. Larger PV grid-connected system can occupy a significant area.
However, a PV plant can deliver electricity during 25 or more years, almost CO2 emissions
free, contributing in a very positive manner to the carbonization of the electricity sector.

3.2.2 Hydro Resources Potential

Beyond solar energy sources, hydro energy is one of renewable energy sources available
in the country. Timor-Leste is in the tropical regions, faced with high humidity, temperature, and
abundantly rainfall. Timor-Leste experiences two seasons, the rainy and the dry seasons. The rainy
season starts from the month of November to the month of February. While the dry season starts
from March to October. Studies conducted by the Electrification Plan of Timor-Leste and the
Norwegian Cooperation shown that there was potential to install hydropower with the total
capacity around 380 MW with the annual average of energy production approximately 1250 GWh.
There are some places identified that have the potential for hydropower plant projects, such as in
the Ira Lalaro, Gleno, Belulic, and Laclo. The most viable site, based on the minimum cost of
generation projection, was the 28 MW Ira Lalaro hydro project in the eastern part of the country.
The Ira Lalaro lake is able to provide water naturally for the plant. Likewise, the study also shown

22
that, micro and mini hydro are likely to have an important role in electrification of rural villages
(Martifer, 2010).

Environmental Impact.
Environmental impact associated with generating electricity from hydro energy technologies
are not serious measurable because hydropower does not pollute the water and the air. Even
though, the development of hydropower generation projects can arise many issues such as
environmental authorization, settlement of the affected community and land asset.
On the other hand, hydropower is a mature and affordable technology that can produce
electricity for many years, more than 50, being a very valuable contribution to mitigate gas
emissions and other inconvenient of the use of fossil fuels. It must also be considered that water
can easily be stored, that´s why hydroelectricity can provide services to the electrical system that
other renewables cannot do, as a consequence of its intermittency.

3.2.3 Wind Energy Resources Potential

Wind energy is the energy coming from the wind, and wind is available almost everywhere.
The velocity of the wind, wind direction and seasonal patterns can change considerably. Wind is
another renewable energy source with considerable future power plant in Timor-Leste.
A previous study elaborated by the Electrification Plan of Timor-Leste and the Norwegian
Cooperation shown that annual average wind speed could reach 7 m/s at certain places of the
country. Furthermore, it also found that wind power was not viable in the coastal areas, but it could
prove to be economically viable in the mountains and uplands. Timor-Leste wind atlas is provided
in figure 10. As can be seen from the figure, the mountain areas of eastern parts of Quelicai, the
southwest of Venilale, and east of Maliana have a great potential wind resources compared to the
rest of the territory (Martifer, 2010). An Interesting feature of the wind resources is that the dry
season, from April to October, is also the windy season. There would be a great potential for
complementary with hydropower plant.
Figure 11 shows wind speed values for Dili district, the capital of Timor-Leste, during a
decade, from 2005 to 2015. Data was collected from the Dili International Airport which is very
close to the sea, and the terrain is flat. It is about six kilometres, to the east, from the Hera Campus.
The wind speed at 3 meters from the ground was about 10 to 13.6 kilometres per hour. The highest

23
and lowest wind speed appeared in the month of August and April, and accounted for about 3.8
and 2.8 m/s, respectively.

Figure 10. Timor-Leste wind atlas.

SOURCE: MARTIFER Report, p. 118.

The average wind speed was about 3.2 m/s (Timor-Leste Government, Office of Direção Nacional
de Meteorologia e Geofisica, 2018). Based on the obtained values, it is definitely not a suitable
site to install a wind turbine, but it can be installed a small wind turbine to combine with solar, and
diesel generator to serve a small load.

24
 Figure 11. Average wind speed for Dili, Timor-Leste.

Environmental Impact.
The mainly concern about the impact of installing wind turbines is noise pollution. As turbine
rotates, it generates aerodynamic noise, generally from the turbine blades. According to the
Ministry of New and Renewable Energy (MNRE), one single turbine at a distance of 40 meters
can produce noise about 50 to 60 db (2013).
However, as wind speed values in the uplands are suitable for installing a small wind
turbine to serve low electrical demand in the area that are far from the main grid. Beyond that, this
technology is available in the market and with various capacity that can be chose based on local
energy demand levels.

3.2.4 Biomass Energy Resources Potential

A renewable resource of energy other than solar energy sources, hydropower and wind is
the biomass. Biomass is conferred as the organic materials derived from vegetables, animals’ dung,
agriculture waste, industrial and municipal waste that can be recycled for energy production. It has
been considered as renewable and carbon-neutral fuel as it has great potential of replacing as an
alternative energy source that can address the global energy issues (Tripathi et al. 2016). Biomass
gasification is an effective technology for decentralized electricity generation from agriculture
waste and wood residues. The development of this type of technology has revealed to be relevant

25
in terms of growth of local energy services, efficiency and economics benefits trough job creation
(Chidikofan et al., 2017).
The Electrification Plan of Timor-Leste and the Norwegian Cooperation carried out a study
on analyzing the feasibility of the amount of biomass residues that can be converted into useful
energy in thirteen districts across the country. The study results can be viewed in table 1. As can
be observed, agro-forestall residues have the highest quantities compared to the agriculture and
animal residues. The study has also shown that there is a potential to generate electricity from
Dili’s Municipal Solid Waste (DMSW) building a solid waste recovery plant. The author estimates
that a 3 MW capacity of DMSW energy recovery plant could be developed. The annual energy
production was estimated about 24.1 GWh (Martifer, 2010).

Table 1. Distribution of usable biomass residues by district.

Agriculture Agroforestry Animals
District
(tons/year)
Aileu 507 236,598 1,273
Ainaro 1,990 212,986 4,883
Baucau 4,116 401,867 7,587
Bobonaro 8,840 432,679 6,347
Covalima 3,446 501,992 4,436
Dili 312 128,226 4,133
Ermera 2,147 253,375 3,147
Lautem 11,284 619,409 3,369
Liquica 735 168,593 4,562
Manatuto 3,718 741,822 3,619
Manufahi 3,217 369,768 3,042
Oecusse 2,979 161,062 5,104
Viqueque 5,897 673,106 9,134
Total 49,186 4,901,484 60,637

SOURCE: MARTIFER Report, p. 87

Environmental Impact
The concern about the impact of biomass generation is the sulfur oxide and nitrogen oxide
emissions. Burning of biomass can also produce carbon dioxide which contributes to the GHG
emissions. In addition to GHG emissions, biomass power generation uses water in the boiler for
cooling system. The temperature of the water cooling coming out from the exhaust boiler is high,
thus can harm plants and fish in the river or lake where the biomass power plant water is
discharged.

26
4. Methodology

In order to answer the key questions, set out in this dissertation, three main specific tasks
were carried out. The first task was to identify the location and the description of the system that
will be studied. The second task was to determine the performance of the PV system including
reference yield which expresses an equivalent number of hours at the reference irradiance,
following with evaluating the final yield which represents the net electricity produced, divided by
the rated power output. Also, load profile and power generate from the system were analyzed and
the potential for expanding the system and ascertain the potential for battery storage were assessed
in this section. A performance ratio was then analyzed to determine how close the PV system
approaches the ideal performance under actual operating conditions. Next, energy losses and
system capacity factor were evaluated. Finally, the efficiency of the system was analyzed to
determine how satisfactory is the performance of the PV system components. The third task was
to use HOMER Pro software to analyze the amount of energy that could be stored into a battery
when it is attached to the system in order to meet the required load demand. A sensitivity analysis
was also performed to comprehend the current PV system capacity in responding to the load
fluctuations.

4.1 The PV System Description

The major elements of a PV grid-connected system are PV arrays, junction box, collecting
box, inverters, transformer, and commercial grid. PV modules generate DC power. This DC power
is the product of voltage and current. The power conditioner converts DC voltage to AC voltage.
The AC output from the power conditioner is then supply to the building and main grids. For a
residential use, the AC output is supplied to the buildings and the excess of the electricity is sold
to the utility. In contrast, for a commercial use, the AC output of power conditioner is supplied
directly to the main grid through utility meters. In PV grid-connected system, the inverter always
operates in phase with the grid phase.

27
 Figure 12. PV grid-connected system at FEST.

The Faculty of Engineering, Science, and Technology of the Universidade Nacional Timor
Lorosa’e installed a 250 killowat PV Grid-connected System in 2014. The latitude and longitude
of the site are 8033.1’S and 125039.6E, respectively. The system, as shown in figure 12, consists
of 1200 PV panels, polycrystalline silicon, of 215 W arranged in 12 series-connected modules.
The PV power plant consists of five arrays. Each array is made up of four panels. The system is
mounted with the inclination of 10 degrees, as shown in figure 13, facing North (Office of Faculty
of Engineering, Science, and Technology, 2018).

Figure 13. PV plant mounting inclination.

28
The technical specifications of the PV panels are provided in table 2.

Table 2. Technical specification of PV module Kyocera KD215GH-2PB.

Electrical data
Maximum power, Pmax 215 W
Maximum power voltage, Vpm 26.6 V
Maximum power current, Ipm 8.09 A
Open circuit voltage, Voc 33.2 V
Short circuit current, Isc 8.78 A
Maximum system voltage 1000 V
Module efficiency 14.4%
Limitting reverse current (series fuse rating) 15 A
Reduction of efficiency (from 1000 W/m^2 to 20 W/m^2) 6.0%
Nominal Operating Cell Temperature,NOCT 45_degree C
Temperature properties: Temperature coefficient
V_OC [V/℃](V_OC [%/℃]) -1.2*10^-1*(-0.36)
I_SC [A/℃](I_SC [%/℃]) 5.27*10^3(6.0*10^-2)
P_max [W/℃](P_max [%/℃]) -9.91*10^-1(-0.46)
Cell temperature 25 C
Standard Test Condition (STC)
Spectrum AM 1.5; Irradiance level 1 kW/m^2

The system is connected to two 100 kW inverters of P83BR104R model, and one 50 kW of PMC
500 model. A 350-kVA transformer is used to distribute electricity to the commercial grid and
Faculty buildings.

4.2 Parameters Used in the Analysis of the Performance of the PV System

The performance parameters used in this study are the ones developed by International
Energy Agency (IEA). The parameters were developed for analyze the performance of grid-
connected PV system. There are several parameters used to define the overall system performance
regarding the energy production, solar resource, overall system losses, performance ratio, capacity
factor, PV array efficiency, and PV system efficiency. For this study, some derived parameters
related to energy and performance were calculated using the monitoring data recorded.

4.2.1 Reference Yield

The reference yield (YR) is the ratio of the incident energy in the array plane HI to the array
reference irradiance GO. The total daily in plane irradiation is in units of kWh/(m2.day) and
reference irradiation is equal to one kilowatt per square meter. So, the reference yield is in unit of

29
kWh/kW/day or in hours per day. It expresses an equivalent number of hours per day during which
the solar radiation would necessary to be at reference irradiance levels in order to contribute with
the same incidence energy as was monitored. It can be calculated using equation (Kumar et al.,
2017):

𝑯𝑰
𝒀𝑹 = (4.1)
𝑮𝑶

Where:
HI = the incident energy in the array plane (kWh/m2/day)
GO = array reference irradiance = 1 kW/m2

4.2.2 Final Yield

The final yield (YF) of a PV system for a given period represents the net AC energy output
divided by the rated power of the installed PV array at standard test condition (STC) of 1000 W/m2
solar radiation and 25 0C cell temperature. It indicates the number of hours of operation of the
array required per day at its rated capacity to equal its monitored contribution to the daily useful
energy. It can be written as (Kumar and Shudakar, 2015):

𝑬𝑨𝑪
𝒀𝑭 = (4.2)
𝑷𝒎𝒂𝒙𝑮,𝑺𝑻𝑪

Where:
EAC = energy produced by the PV system (kWh)
PmaxG,STC = PV array rated power at standard condition test (STC) (kW)

4.2.3 Performance Ratio

The performance ratio (PR) is an indicator that normalizes the energy supplied to the grid
with respect to nominal power mentioned in the specification of the PV array. Also, it indicates
the overall effect of losses on a PV array´s normal output power. Its values indicate how close a
PV system approaches ideal performance under actual operating circumstances. The PR is equal
to the final yield divided by the reference yield. It can be formulated as (Kumar et al., 2017):

30
 𝒀𝑭
𝑷𝑹 = (4.3)
𝒀𝑹

Where:
YF = the final yield of a PV system (h/d)
YR = the reference yield (h/d)

4.2.4 Total Energy Losses

The total energy losses (LT) are the PV losses due to irradiance level, array temperature,
module quality, ohmic wiring, mismatch and inverter losses. It represents, numerically, the
difference between the reference yield and the final yield. This can be expressed by the following
equation (Attari et al., 2016):

𝑳𝑻 = 𝒀𝑹 − 𝒀𝑭 (4.4)

4.2.5 Annual Capacity Factor

Annual Capacity factor (CFA) is defined as the ratio of the real annual energy output from
the plant to the amount of energy the system would produce if it worked at full rated power for 24
h/day for a year (365 days). The CF has a direct implication on the cost of electricity production.
The annual value of the CF is calculated by taking cumulative sum of useful energy values at one-
hour intervals using monitoring data recorded. It can be expresses as (Allouhi et al., 2016):

𝑬𝑨𝑪,𝒀𝒆𝒂𝒓
𝑪𝑭𝑨 = (4.5)
(𝑷𝒐 ∗𝟐𝟒∗𝟑𝟔𝟓)

Where,
EAC,Year = annual energy produced by the PV system (kWh).
PO = rated power output of the installed array (kW).

4.2.6 System Efficiency

The efficiencies of the PV grid-connected system components are array mean efficiency,
inverter efficiency and overall system efficiency. Array mean efficiency (PV,A) describes the ratio
of the annual energy output of the system to the total energy collected from the PV field (Allouhi
et al., 2016). It is given by equation (4.6) in which, EAC,year is the energy output of the inverter, and

31
APV is the area of the PV modules. The inverter efficiency is given by equation (4.7) in which, EAC
is the energy output of the inverter, and EDC is the DC energy input of the inverter considered over
the reporting period. The overall system efficiency is given by equation (4.8) (Padmavathi and
Daniel, 2013).

𝐸𝐴𝐶,𝑌𝑒𝑎𝑟
𝑃𝑉,𝐴 = (4.6)
𝐻𝐼 ∗𝐴𝑃𝑉
𝐸𝑂𝑢𝑡
𝜂𝐼𝑛𝑣 = (4.7)
𝐸𝐼𝑛

𝜂𝑆𝑦𝑠 = 𝜂𝑃𝑉 ∗ 𝜂𝐼𝑛𝑣 (4.8)

4.3 Amount of Energy Stored in the Battery when Connected to the System

Computer simulation allows optimization of different economic and engineering

parameters that need to be considered in order to design and construct an energy system. It can be
used to present a feasibility study of a new system, and also to analyze problems that can arise in
the system´s operation. The development of a simulation software enables the designer to diagnose
in order to discover the most acceptable level of a renewable energy system. HOMER Pro software
is chosen in this study as it is easy to configure a system. The HOMER Pro simulation tools allow
for the evaluation of a range technology combination over different constraints and sensitivity
inputs to optimize energy system. This software helps to determine the best scenario that combines
grid and renewable power, storage, and load management to ensure a consistent and reliable
microgrid.

4.3.1 HOMER Pro Input Model

HOMER Pro simulation model is illustrated in figure 14. It can be observed that neither
energy generated by the PV system and grid can directly supply energy to meet the FEST energy
demands. In the case that there is any surplus of energy generated, it is fed directly to the utility
grid. Some of the energy generated can be stored in the battery attached to the system. The
electrical load, solar radiation and the ambient temperature input for the simulation were all values
taken from the average values of data from 2015 to 2016. Simulations for each individual year
were also executed in order to compare both energy production and consumption changing.

32
Another parameter input such as PV characteristics, inverter and grid characteristics remain
unchanged.

Figure 14. Simulation model for HOMER Pro.

4.3.2 Primary Load Input

To carry out the simulations and analysis, it is necessary to know the electrical load
requirement of the FEST building. HOMER Pro simulation tools requires hourly load input data
during 24 hours per day for each month. The average diurnal energy consumption from 2015 to
2016 including each individual year were used in this study so that the comparison in PV energy
generated will be easier to show. In addition, different data from those years were used in order to
see the trends and effects of different energy consumption to the simulated battery used. The
electrical load input for the average monthly from 2015 to 2016 can be viewed in figure 15, while
for each individual year can be seen in Annex A2 to A3.

33
 Figure 15. Average electrical load of 2015 and 2016.

4.3.3 PV Input Characteristics

The PV parameters such as nominal operating cell temperature, module efficiency, and
temperature coefficients were entered to the HOMER Pro program. These parameter values are
based on the PV characteristics provide on table 2. Power input for PV array capacity is 250 kW.
The derating factor which is the scaling factor that HOMER applies to the PV array power output
for reduced output in real-world operating condition compared to the conditions under which the
panels was rated. Use the derating factor to account for such factor as soiling of the panels, cable
losses, shading, snow cover, aging and so on were assumed to be 80% (is a typical number) and
the ground reflectance was set to 20%. PV characteristics used for the simulation are shown in
figure 16.

34
 Figure 16. PV input model

4.3.4 Grid Input Characteristics

Grid power price was set to 0.12 US$/kWh according to the current power purchase
agreement in Timor-Leste. Grid sellback was set to “ZERO” kWh because no FIT policy exists.

4.3.5 Inverter Characteristic Input

The inverter capacity and inverter efficiency were set to 250 kW and 90%, respectively.

4.3.6 Solar Radiation Input

Solar radiation data used in this simulation was the monthly average solar radiation data
from 2015 to 2016. HOMER Pro automatically calculates clearness index values when solar
radiation data is entered. Solar radiation data of the average of the given two years period can be
viewed in figure 17, while for 2015 and 2016 can be seen in Annex A5 to A6.

35
 Figure 17. Average solar radiation values of the three years studied periods.

4.3.7 Temperature Input

Monthly average ambient temperature from 2015 to 2016 were used as an input. These
temperature values were shown in table 5. In addition to that, an average temperature from 2015
to 2016 was also entered in the models as well. This average temperature can be viewed in figure
18.

Figure 18. Average temperature of the three years period.

36
4.3.8 Battery Input

The battery types used in the simulation tool were UET-ReFlex and CELLCUBE FB20-
100. A sensitivity analysis will be performed by considering an increasing number of batteries,
that is, different storage capacity of the system. The characteristics of the batteries included in
the system are given in table 3 and table 4.

Table 3. The characteristics of the UET-ReFlex-100 kW battery.

Battery Specification
Type UET-ReFlex-100 kW

Lifetime 20 Years

Peak Power 120 kWAC

Maximum energy 450 kWhAC

Available State of Charge 100%

Efficiency peak Shaving 75%AC

Efficiency frequency regulation 70%AC

Voltage range 400AC-10% to 480AC+10%

Ambient temperature -400C to 500C (-400F to 1220F)

Table 4. The characteristics of the CELLCUBE FB20-100 battery.

Battery Specification
Type CELLCUBE FB20-100
Lifetime 20 Years
Nominal Capacity 100 kWh
Roundtrip efficiency 64%
Maximum discharge current 598.958 A
Output volatge 400 VAC

In this study, the minimum allowable state of charge, SOC, is 50%, the initial SOC is 100%, and
assuming that charge and discharge occurs every day. Also, storage may be useful for buffering
the electric grid when there is a sudden change of solar irradiation. The input variable of round-
trip efficiency, is the round-trip DC-to-storage-to-DC energy efficiency of the battery bank or the
fraction of energy put into the storage that can be retrieved. HOMER assumes that the storage
charge efficiency and the storage discharge efficiency are both equal to the square root of the

37
round-trip efficiency. Typical round-trip efficiency is about 80%. However, as both batteries have
their own round-trip efficiency then those numbers were used to run in the simulation.

4.3.9 Control Unit

Control unit monitors the power monitoring unit of the system and delivers power based
on the scenario: “The PV delivers the required energy to the building and store the excess energy
in the battery. Battery supplies the required energy when there is insufficient solar radiation until
it gets its Depth-of-Discharge, DoD”.

4.4 Sensitivity Analysis

Sensitivity analysis was performed based on considering the uncertain parameters and
removing the unrealistic combinations. Sensitivity analysis is the study of the sensitivity of a
system when the parameters change their values (Salmani et al., 2014). HOMER Pro shows how
the effect of the system changes with the fluctuation in load demand, solar radiation and ambient
temperature. In this study, different sensitivity variables were considered to evaluate how the
current PV system can meet the fluctuation of energy demand. In addition to that, a sensitivity
analysis of PV system with battery and load fluctuation was also conducted. The fluctuation load
demand considered in this sensitivity analysis were load demand increases by 50%, 100%, 150%,
and 200%.

38
5. Results and Discussion

5.1 Solar Radiation and Ambient Temperature of the Site

Timor-Leste is located in the tropical region which have high humidity and temperature
with a significant amount of rainfall, and cloudy sky affects the solar radiation incident on the PV
system. In situation of cloudy weather, direct sunlight is scattered by cloud particles in atmosphere
and hits PV module as diffused light thus affects the PV efficiency.
The average monthly value of solar radiation and the ambient temperature for 2015 and
2016 for the FEST of UNTL is depicted in table 5. The data was collected from the monitoring
rooms of FEST of UNTL. The data recorded include the maximum, minimum, total and average
values of all parameters over one-hour periods. Highest average solar radiation was about
6 kWh/m2 and occurred in the months of August and September. On the other hand, the ambient
temperature, was found to be higher during the month of November and it was lower during the
month of July and August. Overall, annual average solar radiation and ambient temperature were
about 5.35 kWh/m2 and 26.57 0C, respectively.

Table 5. Average monthly solar radiation and ambient temperature for FEST of UNTL.

39
5.2 Electrical Load Profile for FEST of UNTL

The FEST opens from 8.30 AM to 17.30 PM from Monday to Friday. Normally,
educational activity starts from 9.00 AM and ends by 17.00 PM. At this time range, all equipment
in all departments are to be ON for an average period of 9 hours. The total monthly electrical loads
for the FEST during 2015 to 2016 and the average of the both periods are given in figure 19.

Figure 19. Total monthly energy consumption for the FEST of UNTL.

It can be seen from figure 19 that more electricity was used in 2015 compared to that of
2016. In 2015, the higher energy consumption occurred in the month of March while the lowest
happened in December, accounting for about 23.13 and 12.83 MWh, respectively. This is because,
in March, more laboratory activities were conducted, while December was the end of the
academic year in Timor-Leste, so there were no more student’s laboratory activities. Total energy
consumed in 2015 was about 232.42 MWh. Similarly, monthly load values for 2016 was higher in
March as well, accounted for about 19.90 MWh, and was less for July, accounted for 10.35 MWh
with a total load of 178.60 MWh. The total annual average of energy consumption for both
years was about 205.51 MWh. It can be observed that energy consumption in 2016 was reduced
by 13% compared to the previous year. This phenomenon happened as FEST reduced its

40
laboratory activities due to a financial issue (Faculty of Engineering, Science and Technology,
2018).

5.3 Diurnal Average Energy Generated and Energy Consumption

The average diurnal energy consumption of the FEST and energy generation of the PV
system from 2015 to 2016 are presented in figure 20. The graph shows hourly monthly average
energy demand and energy produce from the PV system. This trend represents the daily energy
production and consumption from the system during the 24 hour per day per year.

Figure 20. Diurnal average energy consumption of the FEST and energy generated by the PV system.

From figure 20, the average daily maximum energy consumption for the two years period
was just about 56 kWh which occurred during the mid-day time. The minimum load consumption
happened after learning activities finished and remained almost constant throughout the night until
early in the morning; it accounted for around 7 kWh on average. The average monthly energy
consumption was 565 kWh. On the other hand, monthly average energy produced by the PV
modules was about 892 kWh.
Beyond the average values of energy generated by the PV modules and energy consumed
by the FEST, the diurnal average energy generated by the system and FEST energy demand for
each individual year were also presented. The average of daily energy uses trends for 2015 and

41
2016 were similar to the one presented in figure 20. The result can be seen in annex G2 and G3.
The monthly average energy produced and load consumed in 2015 were about 960.67 and 639.71
kWh, respectively. The amount of energy consumed during the night until early in morning was
about 9 kWh on average. As for 2016, the monthly average energy produced and load consumed
were about 850.13 kWh and 488.31 kWh, respectively. The amount of energy consumed during
the night until early in morning was about 5 kWh on average.
It can be observed that the majority of energy use in the engineering buildings is due to the
laboratory activities. FEST laboratory activities start at around 10.00 AM to about 16.00 PM local
time (Office of Faculty of Engineering, Science, and Technology. 2018). Regarding this time
schedule of the engineering activities, the option for a PV system is favorable. This is because
during this period energy from the sun is abundant. The study results show that the energy
generated during that interval, from around 9 AM to 16 PM, is higher than that of energy
consumed.
The study results also reveal that energy generated from the system was higher than energy
consumed for both periods. This proves that energy generated was more than sufficient to cover
the load demand and hence does not require to expand the system. Nevertheless, there is a potential
to integrate battery storage with the system in order to store the excess of energy generated. Storing
this excess energy in storage batteries permits it to be used to supply electrical load during night
time or when the power from the grid is interrupted. Economically, install battery would require
high capital cost but there are several technical benefits, such as communication benefit. For
instance: student can still access to the internet when power from the main grid is interrupted.
Another technical benefit is that the learning process continues to function whenever there is an
interruption from the grid. Moreover, the FEST of UNTL can be categorized as self-sufficient
energy consumption because during the day time, power is provided by the PV system while
battery offers power at night time. This kind of benefits must be taken into consideration when
decision is made. In addition, it is one of the measures that can be used to secure the continuity of
power supply to the FEST buildings and to improve the reliability of the Hera PV system.

42
5.4 Energy Generated and Energy Injected to the Grid by the PV System

Figure 21 shows the total amount of energy generated by the PV plant. It can be spotted
that PV modules produced more energy in 2015 compared to that of 2016. In 2015, the monthly
energy produced ranged between a minimum of 23.32 MWh, in December, and a maximum of
35.36 MWh in September. For 2016, the energy generated values ranged from 16.21 MWh, in
February, to 34.76 MWh in October. The annual average of total energy generated by the PV
modules during the two years was around 324.94 MWh.

Figure 21. Energy generated by the PV system.

It can also be observed from the graph that energy generated in February, May to July, and
December in 2016 was lower than values obtained in 2015. One aspect that contributed to the
lower energy generated values is lower output from the PV modules due to higher cell temperature
beyond 25 oC. This higher cell temperature is caused by the higher ambient temperature, Ta. This
higher cell temperature will reduce PV modules voltage operating circuit (Voc) thus reduce the
energy output from the PV modules. Amelia et al. (2016) conducted an investigation on the effect

43
of temperature on PV panel output performance and found that, an increase of temperature of 100C
would decrease the energy output of about 5%.

In terms of the energy exports, as it can be seen from figure 22, the monthly average of
energy exported to the grid were ranged from 6.7 MWh to 19.1 MWh. Total annual average
energy exported to the grid from 2015 to 2016 was about 162.6 MWh.

2015 2016 Average

20000.00
Energy injected to the grid, kWh)

18000.00

16000.00

14000.00

12000.00

10000.00

8000.00

6000.00

4000.00

2000.00

0.00
1 2 3 4 5 6 7 8 9 10 11 12
Month of the year

Figure 22. Energy injected to the grid from the PV system.

As far as I understood, there is no feed-in-tariff and there is no tariff at all. Energy could
be paid at market price. The absence of this regulation has a negative impact for the FEST as a PV
grid utility owner. It causes FEST-UNTL cannot earn economic benefits when the system supplies
the excess of energy produced to the grid. This shows that the sustainability of the operating of the
system cannot be guaranteed. In fact, the FEST has now found financial difficulties to do
maintenance and to replace the broken inverters and PV modules. The absence of the FIT
regulation can also affect the interest of other renewable electricity company to invest in the
country. Accordingly, the renewable energy mechanism should be implemented to crab attention
of investors. Despite renewable energy mechanism, policy such as capital subsidy, tax incentives,
private investment, and foreign direct investment can also be employed to enlarge renewable

44
energy sources. As without these mechanisms, it is very difficult to achieve the national goals and
visions of renewable energy deployment. Nesamalar et al. (2017) stressed that policies and
supports are essential for promoting renewable energy source-based generations.

5.5 Energy Supplied and Imported from the Grid to the FEST

Table 5 shows the monthly values of energy supplied by the PV plant and imported from
the grid to the FEST for the period of 2015 and 2016. It can be observed that total energy supplied
by the system and energy imported from the grid to the FEST was higher in 2015 and was less for
2016. The annual average of total energy shared by the system and brought from the grid were
about 161.46 MWh and 43.18 MWh, respectively.

Table 6. Average energy supply to the FEST buildings.

Energy supply by the PV system and energy import from the grid to FEST buildings
Energy supply by the PV system Energy import from the grid
Month
2015 2016 Average 2015 2016 Average
Jan 11724.30 11912.10 11818.20 4461.56 1895.35 3178.46
Feb 12586.20 8491.07 10538.64 4689.24 4493.99 4591.62
Mar 17095.40 16985.00 17040.20 6030.11 2914.34 4472.23
Apr 15457.10 13861.80 14659.45 5648.27 2679.84 4164.06
May 15320.50 10501.98 12911.24 5764.51 3045.10 4404.81
Jun 14309.70 9798.89 12054.30 5701.82 2935.38 4318.60
Jul 15400.30 8364.53 11882.42 5129.32 1983.71 3556.52
Aug 16123.40 13280.40 14701.90 4841.14 1346.21 3093.68
Sep 17576.00 16290.50 16933.25 4287.91 1989.08 3138.50
Oct 15954.90 15727.10 15841.00 3260.55 1692.21 2476.38
Nov 12636.10 14039.30 13337.70 3874.25 3315.65 3594.95
Dec 10581.30 8900.99 9741.15 2238.59 2139.19 2188.89
Total 174765.20 148153.66 161459.43 55927.27 30430.05 43178.66

The overall amount of energy consumption, energy produced by the PV modules, energy
exported and imported from the grid can be seen in table 6.

Table 7. Total energy production and consumption by the FEST

Year/Energies Ep, MWh Ee, MWh Percentages, % Ec, MWh Ei, MWh Percentages, %
2015 349.20 172.73 49.46% 232.42 55.93 24.06%
2016 300.68 152.53 50.73% 178.59 30.43 17.04%
Average 324.94 162.63 50.05% 205.51 43.18 21.01%

45
Where:
Ep = Energy produces by the PV modules (kWh)
Ee = Energy exports to the grid (kWh)
Ec = Energy consumptions of FEST (kWh)
Ei = Energy imports from the grid (kWh)

It can be seen that the system exported about half of its total energy generated to the grid,
meanwhile the FEST imports less than 25% of energy from the grid to satisfy its energy demands.
These values mean that the system injected a 50% of its total energy generated into the grid for
free, while must pay for the 25% of energy that are imported from the grid. In order to minimize
energy imports from the grid, attaching a battery to the system is a favorable option. By doing this,
FEST will have sufficient energy supply from PV system and there is still excess energy supply to
the grid for free.

5.6 System Reference Yield, Final Yield and Performance Ratio

To analyze the performance of the Hera PV Grid-connected system, data monitoring from
2015 to 2016 was used. The estimation was carried out by calculate relevant parameters for the
average values. The calculation for the two years performed individually in order to compare the
PV performance levels. The calculated parameters were the YF, the YR, and the PR. Operating
results regarding to two years are presented in the next sections. The average YR, YF, and PR from
2015 to 2016 can be viewed in figure 23.
Figure 23 shows the annual average YR, YF, and PR trends from 2015 to 2016. It can be
observed that YR ranged between a minimum of 4.39 kWh/(kWp.d), in December, and a
maximum of 6.23 kWh/(kWp.d), in September. The annual average values o f YF ranged
from 2.49 to 4.52 kWh/(kWp.d). It also shows that the PR ranged from 57%, in December, to 73%
in September. These values were very stable which indicated that the PV system has operated
correctly.

46
 Yr Yf PR
7.00 0.80
Yr and Yf, kWh/(kWp.d)
6.00 0.70

0.60
5.00
0.50
4.00

PR
0.40
3.00
0.30
2.00
0.20
1.00 0.10

0.00 0.00
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month of the year

Figure 23. The annual average values of Yr, Yf, and PR.

However, the calculation of each individual year of operation were also carried on in order
to see the trends of the system performance under real conditions. The study results show that, see
annex E6, the monthly average values of PR for 2015 was less in November, accounting for about
59.63%, and was higher in January, accounting for around 80.13% with an annual average of
70.23%. The annual average value of YR and YF were around 5.31 and 3.71 kWh/(kWp.d),
respectively. On the other hand, the annual average PR for 2016 was about 58.47%. While the
monthly average values ranged between a maximum of 75.83%, in March, and a minimum of
42.13%, in July. The annual average value of YR and YF was 5.39 and 3.19 kWh/(kWp.d),
respectively.
The study results reveal that the PR values of the system at the first phase of operation were
higher. This is because the energy output, which represent by YF, increases with a good amount of
solar radiation and a constant ambient temperature level. On the other hand, as for 2016, the PR
was less due to higher ambient temperature. This higher ambient temperature reduced energy
output, YF, of the PV modules slightly even there was a slightly increased of solar radiation. The
trend can be clearly observed from table of summary of performance parameters values in annex
E6.

47
 The YF of other monitored system include: Morocco ranged from 1.96 to 6.42
kWh/(kWp.d) (Attari et al., 2016), India from 3.56 to 5.09 kWh/(kWp.d) (Kumar et al., 2015), and
Greek from 1.96 to 5.07 kWh/(kWp.d) (Al-Otaibi et al., 2015). The average PR of other observed
PV plant varied from country to country with an average value ranging from 49% up to 98%. A
similar study conducted in Ghana found the PR ranges between 49% and 71% (Quansah et al.,
2017); India, 80% (Kumar et al., 2017); in Serbia found the system PR was about 94%
(Milosavljevic et al., 2015). In the similar study conducted in Morocco found the PR values was
ranged from 58% to 98% (Attari et al., 2016), this trend shows the highest PR value among other
observed countries. When comparing these PR values with the PR value obtained from Hera PV
plants at the first-year of its’ operations, the PR obtained was higher than that of in Ghana but
lower than in Morocco, Serbia and India. In contrast, the PR values obtained in 2016 were lower
than those in mentioned countries.

5.7 PV System Total Losses

The average monthly LT values from 2015 to 2016 represented in figure 26 were calculated
theoretically using the formula (4.4). The study results show that the average monthly Ls was
different for each month. The monthly average LT value was high for July, accounted for about
2.62 kWh/(kWp.d), and was less for January accounted for 1.37 kWh/(kWp.d) with an annual
average of 2.04 kWh/(kWp.d). For each individual year, the average monthly LT values of 2015
and 2016 were 1.74 and 2.34 kWh/(kWp.d), respectively. The changing values of both years can
be seen in table of summary of performance parameters values in annex E6.
There are several factors that contributed to the LT. One factor that caused higher Ls is cell
temperature higher than 25 oC. This higher cell temperature is caused by higher ambient
temperature, Ta. This higher cell temperature will lower voltage open circuit (Voc) of the PV
modules. This low Voc reduces the YF, of the PV modules. Once the energy yield by the array
plane, YA, is higher than YF, then LT would be higher. Other factors that contributed to LT are losses
through cables due to the effect of the temperature on all the cables system. These losses are mainly
due to long wiring between the PV system and the technical room. In addition to that, losses
through energy conversion in the inverter also contributed to the system losses.

48
 3.00

2.50

System losses, kWh/(kWp.d)

2.00

1.50

1.00

0.50

0.00
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Ls 1.37 1.79 1.44 2.11 2.56 2.37 2.62 2.35 1.88 1.85 2.16 2.01

Figure 24. Monthly average PV system losses.

5.8 PV System Capacity Factor

The average monthly capacity factor (CF) values from 2015 to 2016 represented in figure 27
were calculated theoretically using the formula (4.5). The study results show that CF of the system
was varies from month to month. The monthly average CF value was high for September,
accounting for about 19.44%, and was less for December, accounting for 10.72% with an annual
average of 14.83%.

Figure 25. Capacity factor of the installed PV system.

49
 The calculation of CF for each individual year was also performed, see annex E4, in order to
compare the average values of CF occurred under the real operating conditions. The study results
show that the annual average CF values for 2015 was higher than that of in 2016, accounted for
around 15.94 and 13.71%, respectively. While the monthly average CF values, in 2015, ranged
from 12.54% to 19.65%, and as for 2016 was between 8.91% and 19.23%.
The study results indicate that monthly CF values varied from month to month. These values
variation was due to the system losses as a result of local climate conditions. The CF values of
others similar studies shows that, according to Kumar and Sudhakar (2015), a CF value varied
from 12.67% to 20.04% based on one-year operation. Overall CF for the Indian PV plants varied
from 12.29 to 18.8%. Attari et al. (2016) conducted a performance investigation on a PV grid-
connected system in Morocco and found the annual CF of the system to be 14.84%. There were
also studies conducted by Makrides et al. (2007) on performance assessment of a fixed PV system,
found the CF annual average value to be 19.4%. Other studies conducted by Gottschalg et al.
(2005) found CF values between 20.8% and 26% for south-facing fixed and dual-axis tracking PV
systems in Cyprus, respectively. To the knowledge of the author, there were no previous
documentation or studies of CF for a fixed PV system in Timor-Leste. However, it can be deducted
that CF value higher than 8.91% and less than 19.65% is a reasonable for a fixed PV system. The
annual average CF value obtained in this study was 14.83%.

5.9 PV Modules Efficiency

Figure 26 shows the average performance of polycrystalline module under real conditions
of operation from 2015 to 2016. It can be observed that the average monthly PV module efficiency
values varied for each month. The monthly average PV module efficiency value was high for
March, accounting for about 11%, and was less for July, accounting for 8% with an annual average
of 9.33%. For each individual year as it can be seen in table of summary of performance parameters
values in annex E6, the average monthly PV module efficiency values of 2015 and 2016 were
10.17 and 8.47%, respectively.

50
 Figure 26. Monthly average PV module efficiency.

These trends of the PV array efficiency are due to lower PV array output. One aspect that
could lower the PV array efficiency was high ambient temperature. As the ambient temperature
higher, the cell temperature will increase which causes the voltage open circuits of the PV modules
drops. Consequently, energy generated from the PV array will be low. Beyond that, efficiency is
also affected by the energy conversion losses in the inverter and system losses. Therefore, the
efficiency of the PV modules was lower compared to the levels provided by the manufacturer.

5.10 PV System Efficiency

Table 8 displays inverter and system efficiency under factual conditions of operation. It can
be observed that the PV plant efficiency values were different from 2015 to 2016. The average
monthly inverter and PV system efficiency was high for 2015, and was less for 2016. This is
because the efficiency of the PV modules and the inverter efficiency decreased. This reduction
was due to less energy output from the PV arrays which reduce the PV array efficiency.

51
 Table 8. Annual average of Inverter and PV system efficiency.

Year hINV hPV

2015 70.61% 7.23%
2016 58.78% 5.21
Average 64.80% 6.05%

The average monthly values for each individual year can be seen in table of summary of
performance parameters values in annex E6, as it can be noticed that the average monthly PV
system efficiency values of 2015 ranged from 5.18 to 9.35 %. While the inverter efficiency ranged
from 59.95 to 80.57%. As for 2016, the maximum inverter efficiency was 76.25% and minimum
of 42.35%.

5.11 Simulation Results of PV System with Battery

The simulations of several scenarios incorporating storage batteries in the system are made
through the HOMER Pro software and some results are exposed. These results will determine the
conclusions that we will ascertain for this dissertation. In this section, simulation results are
provided based on the average values from 2015 to 2016. Simulation based on yearly data were
also conducted to compare and evaluate the effects of load fluctuation to the simulated battery
capacity.

5.11.1 Simulation Result for the Average data from 2015 to 2016

The overall results of simulation of PV system plus batteries with a different characteristic
are shown in table 9. The table shows, according to each case simulated, the type of battery as well
as the capacity and the energy that has to be purchased from the grid. The overall simulation results
with HOMER Pro software can be seen in annex F1 and F2.

52
 Table 9. Summary results of simulation of PV system with batteries.

Batteries System PV Grid

System,
Capacity, Autonomy, Annual throughput, Ren Frac, Energy Production, Energy Consumption, Energy generation, Energy Purcharged, Energy Sold,
S Type of batteries N batteries
kWh_AC hr kWh % kWh/year kWh/year kWh kWh kWh
4 51 59051 100 365776.6 313746.1 365249 527 107554
S1 UET ReFlex -100kW 2 450 25 53889 99 369762.9 318937.1 365249 4514 112745
1 13 47576 97 374636.5 325247.1 365249 9387 119055
6 25 56861 98 370431 311552.5 365249 5182 105360
5 21 55158 98 371657.3 313356.2 365249 6408 107164
4 17 52982 97 373224.3 315692.1 365249 7975 109500
S2 CELLCUBE FB20-100 100
3 13 49924 97 375425.8 319019.3 365249 10176 112827
2 8.5 45913 96 378313.9 323419.6 365249 13065 117227
1 4.2 32248 93 388152.4 338679.8 365249 22903 132488

Table 9 shows the overall results of PV System connected to batteries with the load
demand, solar radiation and temperature data based on the average values from 2015 to 2016.
Battery of two different types and capacities were chosen to run this simulation. The UET ReFlex-
100 kW type was considered to be system S1, and CELLCUBE FB 20-100 was considered to be
system S2.
The simulation results show that the optimum choice selected by the HOMER Pro for
system S1 was 4 batteries per string. This is because annual battery energy throughput and battery
autonomy are higher compared to the system with 2 batteries and a single one. Beyond that, energy
purchase by S1 with 4 batteries is lower than that of system with a single and 2 batteries. However,
energy production by system S1 in all cases is higher than energy consumption. Also, autonomy
of 1 battery and 2 batteries is sufficient to serve FEST load during grid interruption and even can
serve loads at night. So, the optimum choice of battery could be a system with a single battery.
As for S2, HOMER Pro selects S2 with 6 batteries per string as the best option. This is
because the autonomy of 6 batteries per string is higher than those below 6. The system energy
production of all batteries is higher than the FEST energy consumption. However, as the battery
capacity is quite low, it is useful to look at the battery with a reasonable number of days of
autonomy to decide a battery size. From the result obtained from the simulation, it is possible to
decide that for system S2, the right choice could be system with 4 batteries per string.

53
5.11.2 Simulation Result for 2015 data with Battery

The simulation results for 2015 with batteries can be seen in table 10. As for system S1,
the HOMER Pro selects 4 batteries per string as the most convenient option. However, considering
battery autonomy and energy generated by the system, a single battery and 2 batteries need to be
considered as well. In such case that battery with autonomy of one day is considered and the energy
consumption is assumed to be constant, then option could be the system with 2 batteries per string.
But, if the assumption is made upon the energy generated by the PV system due to abundant
sunlight during the day, then a single battery could be a preferable option.
Considering system S2, the simulation results led to a 6 batteries per string as the best
option. In such case that solar fraction and the autonomy criteria of the system of one day are
considered, then option could be the one selected by HOMER Pro. Nevertheless, if it is considered
that there is more solar production during the day and the energy stored in the battery would be
used only during the night, then a system with 4 batteries per string could be more than sufficient.

5.11.3 The Simulation Results for 2016 data with Battery

The simulation results for 2016 with batteries are shown in table 10. Considering system
S1, the simulation results show that HOMER Pro selects a system with 4 batteries per string as the
preferable option. In such case that FEST would like to independent from the grid, then HOMER
Pro selection can be accepted. This is because the system can cover FEST energy demand at about
100% which make the FEST become energy self-sufficient. However, in such case that FEST
would like contributes some of system energy generated to the grid then a system with single and
2 batteries could be a preferable choice.

As for system S2, HOMER Pro selects a system with 6 batteries per string as the optimal
option. In the case that autonomy of one day is considered then option can be as the same at
HOMER Pro suggested. Nevertheless, in such case that energy production and energy
consumption is considered then the optimal option could be a system with single and 2 batteries
per string.

54
 Table 10. Summary simulation of 2015 and 2016 data of PV system with batteries.

System, Batteries System PV Grid

Year Capacity, Autonomy, Annual throughput, Ren Frac, Energy Production, Energy Consumption, Energy generation, Energy Purcharged, Energy Sold,
S Type of batteries N batteries
kWh_AC hr kWh % kWh/year kWh/year kWh kWh kWh
4 45 74646 99 366744.4 310016.1 364370 2374 76628
S1 UET ReFlex-100kW 2 450 22 68222 98 371704.5 316755.4 364370 7335 83367
1 11 59002 92 378823.2 326357.3 364370 14453 92969
2015
6 23 71359 97 373000.9 308115.8 364370 8631 74728
S2 CELLCUBE FB20-100 4 100 15 65583 96 377159.9 314612.0 364370 12790 81224
2 7.5 54051 94 385462.1 327360.1 364370 21093 93972

4 58 40224 100 361533.8 314401.9 361218 315 135638

S1 UET ReFlex 100kW 2 450 29 38941 99 362524.5 315748.1 361218 1306 136984
1 15 34928 99 365623.3 319958.3 361218 4405 141194
2016
6 29 41331 99 362832.9 310146.8 361218 1614 131383
S2 CELLCUBE FB 20-100 4 100 20 38966 99 364536.3 312808.5 361218 3318 134045
2 9.8 33606 98 368395.0 318837.7 361218 7177 140074

5.12 The Sensitivity Analysis Result

For the sensitivity analysis, the parameters consider, such as climate data and electrical
load are based on 2015 data. The reason for chosen this particular period is due to its high load
demand compared to 2016. Therefore, it is assumed that these data are more representative to run
in this simulation tools. The fluctuation load demand considered in this sensitivity analysis are
load demand increases by 50%, 100%, 150%, and 200%. This percentages are taken based on the
real condition that Faculty´s load demand will increase because currently the FEST is under
progressing of constructions. The FEST building progressing can be seen in the figure 27.

Figure 27. FEST building construction development.

55
Sensitivity Analysis of PV System only with Load Variation Results
The simulation results of the current system capacity with the current average annual
energy demand are presented. Beyond that, the simulations of current system capacity with load
fluctuation scenario are also executed. Attention is given to the energy generation from the
current PV modules in response to any load increase from the current load’s levels.

Scenario 1: Current PV Capacity and Energy Demand

Figure 28 shows the energy output from the system which represented by x-axis, and
according to the time which represented by primary y-axis and secondary y-axis represented PV
power output.

Figure 28. Energy output of the PV modules.

PV modules operate 4015 hours per year with a mean output of 40.93 kW and a maximum
output of 207 kW. The system produces about 358.51 MWh/year with a solar penetration of
85.53%. This value signifies that almost 85.53% of the energy demand has the ability to be met
by solar energy. As it can be observed the energy output is stronger in the middle of the day during
the whole year. However, the PV modules produce less energy often in rainy season due to frequent
rains.

Monthly Power Generation

Figure 29 represents the total power production for each month of the year, including power
production by the PV system and grid purchases. In the case of peak power production, there is an

56
excess of electricity which is not used to meet the load and is dumped to grid. On the contrary, in
the event of high demand and less PV output, electricity is imported from the grid.

Figure 29. PV power production and power imports from the grid.

As it can be noticed the highest electricity import occurs in March while the lowest occurs in
December.

Scenario 2: Current PV Capacity and Increasing Energy Demand

Figure 30 shows the overall sensitivity results of load demand fluctuation that are increased
by 50%, 100%, 150%, and 200% from the current level.

Figure 30. Sensitivity analysis results of PV system plus load fluctuation but without battery.

The sensitivity analysis results, as can be observed from figure 30, of the effect of growths
in energy demand show that the constant increases of energy demand would have impacts on
current system capacity. As it can be seen HOMER Pro selects the preferable option of an increase
load by 1,278.84 kWh/d of a PV energy production value of 358,510.00 kWh/year which account

57
for 64% of total energy generation. Since the PV production occurs during the day time, typically
between 7 am to 6 pm, it is able to meet the energy demand during that time but not in evening.
Therefore, resulting in grid sales of 41,139.00 kWh/year. This is because during mid-day when the
energy production exceeds energy demand, a huge amount of energy is exported to the grid. Total
of 185,257.00 kWh/year of energy is imported from the grid during evening or when the energy
demand exceeds the PV production, this accounts for 36% of the total energy generation. So,
resulting in a total energy production of 543,766.1 kWh/year.
Figure 30 also shows that the addition of load demand results in energy export reduction
and energy import increase. However, as energy import increases higher than energy export when
load demand increases, the renewable fraction decreases. Indeed, 84% of the energy generated by
the current system capacity is consumed for lower load, accounted for 639.42 kWh, while
decreased to about 64% when load demand reaches 1278.78 kWh. It means that the shape of the
load demand appears to be a constraint to the current system capacity because the magnitude of
energy demand depend on the current PV production inevitably results in increasing energy
import. Therefore, HOMMER Pro simulation tools selects an increase load demand up to 100%
from the current levels as the optimal choice.

Scenario 3: Current PV Capacity plus Battery with Increasing Energy Demand

The next scenario is by adding batteries to the system. UET ReFlex-100 kW batteries are
used in this simulation. The battery search space is set to 2 batteries. The overall simulation results
can be seen in annex E8. The results were obtained with a load-following strategy which means
that the batteries are only charged with the excess of PV system output. It can be observed that the
optimal number of batteries is 2, which will be one string with 2 batteries in parallel. The SOC of
the battery is presented in figure 31.

Figure 31. The State of Charge of the UET ReFlex-100kW battery.

58
 It can be noticed from figure 31 that the SOC of the batteries is almost 100% during the
year, because there is an excess of energy that is stored in the batteries during the day. The batteries
annual energy input is 45,121 kWh and the annual energy output is 33,515 kWh. The batteries can
supply the load with an autonomy for 11.16 hours, and the expected battery life is 20 years, where
the renewable fraction is 67%.

Figure 32. The sensitivity analysis results of load fluctuation with PV system plus battery.

In the case that the load is at a constant level of about 639.42 kWh per day, then the PV
system with 2 batteries would provide energy at about 97%, and the number of days of autonomy
is increased to about 22 hours. It means that FEST can be self-sufficient with renewable energy
supply. When energy demand increases to about 200%, then the batteries would only have 7.4
hours of autonomy. Any energy demand increase beyond that level would decrease battery
autonomy and increase energy import. Accordingly, for a system with 2 batteries, the system is
capable to supply energy up to 200% from the current energy demand levels. Having a large energy
demand has effect on the available PV system capacity, because the scale of energy demand
depends on the current system production. Although PV generation varies over days and hours
but only generates energy during day-time and meets energy demand during that period. Therefore,
it is important to control the fluctuation of energy demand that could meet the PV production.

59
6. Conclusions and Recommendations

6.1 Conclusions

The first PV system connected to the grid in the FEST in Timor-Leste was a government
project. In this dissertation a performance analysis was carried out in order to observe the system
behavior under the specific real climate conditions. In addition to that, simulations of the PV
system with batteries were also performed in order to observe how the system can meet load
demand. A sensitivity analysis of load fluctuations was also conducted in order to comprehend the
available capacity with increasing load demand. The simulations of the PV system with batteries
were modelled in the HOMER Pro simulation tool. The HOMER Pro was utilized as the
assessment tool with hourly load data input from the FEST buildings. The input load data and
climate data used were from 2015 to 2016, including the average data of both years.
From the nominal installed power, the average annual energy produced by the system, the
annual energy consumed by the FEST buildings, the Yf, the PR, the CF, and the system efficiency
were estimated. It was found that, the annual average energy consumption by FEST and energy
generated by the PV system from 2015 to 2016 were about 205.51 and 324.94 MWh, respectively.
While energy imported from and exported to the grid were about 43.18 and 162.63 MWh,
respectively. The PR ranged from 55 to about 76% with an annual average of 64%, the annual
average of CF was 14.83% and PV module efficiency was about 9.33%.
As for each individual year, the results of the study show that the total energy generated
was higher for 2015, accounted for 349 MWh, and about 301MWh in 2016. The total FEST energy
consumption was about 232 MWh in 2015 but it reduced to about 179 MWh in 2016. The annual
average value of PR was higher for 2015, about 70%, and about 58% in 2016. The PV modules
efficiency was quite lower for both years compared to the nominal efficiency of 14.4% from the
PV manufacturer.
Simulation that include load fluctuations without battery were conducted. The simulations
results indicate that the system could afford to supply energy up to 100% of the current load
demand levels. Any changes beyond that levels would increase the energy imports from the grid.
While the simulation results of the PV system plus two UET ReFlex-100kW batteries with load
fluctuations indicate the system could afford to supply energy up to 200% from the current load

60
demand levels. However, when load increases beyond that levels, the energy supply from the
system becomes insufficient, the battery autonomy also reduces, and more energy imported from
the grid would be necessary.
In conclusion, after the study carried out in the present work, it is concluded that the
photovoltaic plant installed in the FEST buildings has an acceptable performance. The system
satisfies FEST energy consumption during the day and also injected a vast amount of energy to
the grid. An increase of load in the future would have an impact on the available PV capacity but
it can be managed with a load management technique.

6.2 Recommendations

From the study results, the following recommendation is proposed in order to keep the system
perform well during its operations:
The study results show system output exceeds load demand, resulting in excess of energy
which can be absorbed by adding batteries. Therefore, attaching batteries to the system
should be developed in order to respond to further load increase.
The inverter efficiency is poor, resulting in lower plant efficiency as expected.
Accordingly, it is recommended to replace the inverter to improve the system reliability.
The study results provide a clear information for the government in order to expand and
implements the same project to other government institutional especially for those
institution that has no access to the grid. In addition to that, it provides information to the
top decision maker in the government institutional to select the appropriate types, grid
connected or stand-alone systems, of the system based on energy demand requirements and
local conditions. In the absence of the FIT policy, it is recommended to develop stand-
alone system for further installation.
Allowing other educational institution to access to the system to expand the benefits of this
technology and as well as for academic researcher to conduct research.
Promote the benefits of the system to the students and engaging them in the maintenance
tasks. It would strength their sense of belongings to the system.
Conducting a regular inspection and maintenance to avoid any damages to the system.

61
As for the potential for the use of wind energy, the results obtained indicate that wind speed
at 3 meters from ground is quite low. Therefore, it is recommended for further investigation
of wind speed at higher distances from ground and also at other locations inside, or close
by, the perimeter of the Faculty.

62
7. References

Allouhi A., R. Saadani, T.Kousksou, R. Saidur, A. Jamil, and Rahmoune M. 2016. “Grid-
connected PV system installed on institutional buildings: Technology comparison, energy analysis
and economic performance”. Energy and Buildings 130 (2016):188-201.
doi:10.1016/j.enbuild.2016.08.054.

Al-Otaibi A., Al-Qatan A., Fairouz F., and Al-Mulla A. 2015. “Performance evaluation of
photovoltaic system on Kuwaiti Schools’ rooftop”. Energy conversion and management 95
(2015):110-115. doi:10.1016/j.enconman.2015.02.039.

Amelia, A.R., Iwan, Y.M., Leow, M.Z., Irwanto, M., Safwati, I., and Zhafarina M. 2016.
“Investigation of the effect temperature on Photovoltaic (PV) panel output performance”.
International Journal on Advanced Science Engineering Information Technology. 6 (2016): 682-
688. doi: 10.18517/ijaseit.6.5.938.

Amin,N. Isaka,T. Yamada,A. and Konagai M. 2001. “Highly efficient 1m thick CdTe solar
cells with textured TCOs”. Solar Energy Materials & Solar Cells. 67 (2001): 195-201.
https://ac.els-cdn.com/S0927024800002816/1-s2.0-S0927024800002816-
main.pdf?_tid=27c3f620-14aa-4838-b415-
d3436af2aca8&acdnat=1543368372_174036488f20a15c1fea132704ee54ce.

Attari K., Ali Elyaakoubi, and Asselman A. 2016. “Performance analysis and investigation of a
grid-connected photovoltaic installation in Morocco”. Energy Report 2 (2016):261-266.
doi:10.1016/j.egyr.2016.10.004.

Baş oğlu,M.E., Kazdaloğlu,A., Erfidan,T., Bilgin,M.Z., and Çakır B. 2015. “Performance

analyzes of different photovoltaic module technologies under İzmit, Kocaeli climatic
conditions”. Renewable and Sustainable Energy Reviews 52 (2015):357-365. doi:
10.1016/j.rser.2015.07.108.

Benanti T.L, and Venkataraman D. 2006. “Organic solar cells: An overview focusing on active
layer morphology. Photosynth Research. 87 (2006):73-81. doi: 10.1007/s11120-005-6397-9.

Bruton, T. M. 2002. “General trends about photovoltaics based on crystalline silicon”. Solar
Energy Material ans Solar Cells. 72 (2002): 3-10. doi: 10.1016/S0927-0248(01)00145-3.

Cao H., He W., Mao Y., Lin X., Ishikawa K., Dickerson J.H., and Wayne P. Hess. 2014. “Recent
progress in degradation and stabilization of organic solar cells”. Journal of Power Sources. 264
(2014): 168-183. doi: 10.1016/j.jpowsour.2014.04.080.

Chaar L.E., Lamont L.A., Zein N.E., 2011. “Review of photovoltaic technologies”. Renewable
and Sustainable Energy Review. 15(2011): 2165–2175. doi: 10.1016/j.rser.2011.01.004.

63
Chatterjee,S., Kumar P., and Chatterjee S. 2018. “A techno-commercial review on grid connected
photovoltaic system”. Renewable and Sustainable Energy Reviews 81 (2018):2371-2397. doi:
10.1016/j.rser.2017.06.045.

Chidikofan,G., Benoist A., Sawadogo M., Volle G., Valette J., Coulibaly Y., Pailhes J. and Pinta
F. 2017. “Assessment ofEnvironmental impact of Tar Releases from a Biomasss Gasifier Power
Plant for Decentralized Electricity Generation”. Energy Procedia 118 (2017):158-163.
doi:10.1016/j.egypro.2017.07.034.

Donne, A.Le, Scaccabarozzi, A., Tombolato, S., Binetti, S., Acciarri, M., and Abbotto,A., 2013.
“Solar Photovoltaic: A review”. Reviews in Advanced Sciences and Engineering. 2 (2013): 1-9.
doi: 10.1166/rase.2013.1030.

Ferekides,C. and Britt J. 1994. “CdTe solar cells with efficiency over 15.8%”. Solar Energy
Material and Solar Cells. 35 (1994): 255-262. doi:10.1016/0927-0248(94)90148-1.

Green, M.A. 2006. Third Generation Photovoltaic: Advanced Solar Energy Conversion. Springer,
Berlin.
http://www.bookmetrix.com/detail/book/11d61309-6f65-4756-bffa-91a75440917e#citations

Green M.A., Emery, K., Hishikawa, Y., and W. Warta, 2011. “Solar cell efficiency tables”.
Progress in photovoltaics: Researcs and Applications. 19(2011): 84-92.
https://pdfs.semanticscholar.org/af1f/db800cf316aa7c66343f5db261f14ea41641.pdf.

GreenPeace International. 2011. Solar generation 6: solar photovoltaic electricity empowering

the world, EPIA. Accessed on 28 November 2018. Available at:
http://pvtrin.eu/assets/media/PDF/Publications/Other%20Publications/36.SolarGeneration6__20
11.pdf.

Goetzberger A., LutherL., and Willeke W. 2002. “Solar cells: past, present, future”. Solar Energy
Material & Solar Cells. 74 (2002): 1-11. Accessed on: 27 of November 2018.
https://www.academia.edu/14557148/Solar_Energy_Materials_and_Solar_Cells_74_2002_1_11
_Solar_cells_past_present_future?auto=download.

Gottschalg R, Betts TR, Williams SR, Sauter D, Infield DG, and Kearney MJ. 2005. “The effect
of spectral variations on the performance parameters of single and double junction amorphous
silicon solar cells”. Solar Energy Material and Solar Cells 2005. 85 (2005): 415–428. doi:
10.1016/j.solmat.2004.05.011.

Gul M., Yash K., and Tariq M. 2016. “Review on recent trend of solar photovoltaic technology”.
Energy Exploration & Exploitation. 0(0): 1-42. doi: 10.1177/0144598716650552.

64
Hoeiseth, J. and Klei C. 2007. “Gariaui Mini HEP: The first Hydroelectric Plant in a New
Country”. Paper Presented in the International Conference on Small Hydropower Hydro Sri
Lanka. Sri Lanka, 22-24 October. Assessed on the 5th of June 2018.
http://www.ahec.org.in/links/International%20conference%20on%20SHP%20Kandy%20Srilank
a%20All%20Details/Papers/Policy,%20Investor%20&%20Operational%20Aspects-C/C22.pdf.

Hosenuzzaman M., Rahim N.A., Selvaraj J., Hasanuzzman M., Malek A.B.M.A., and Nahar A.
2015. “Global prospects, progress, policies, and environmental impact of solar photovoltaic power
generation” Renewable and Sustainable Energy Review”. 41 (2015): 284-297. doi:
10.1016/j.rser.2014.08.046.

International Energy Agency, IEA. 2018. “Global Energy & CO2 Status Report 2017”. Accessed
on 30th of May 2018.
https://www.iea.org/publications/freepublications/publication/GECO2017.pdf.

Iles, P.A., 2001. Evolution of space solar cells. Solar Energy Materials and Solar Cells. 68: 1–13.

Kandasamy C.P., Prabu P., and Niruba K. 2013. “Solar potential assessment using PVSYST
software”. In Proceedings of 2013 International Conference on Green Computing, Communication
and Conservation of Energy (ICGCE), Chennai, 2013, 667-672.

Karki P., Adhikary B. and Sherpa K. 2012. “Comparative study of grid-tied photovoltaic (PV)
system in Kathmandu and Berlin using PVsyst”. IEEE International Conference on Sustainable
Energy Technologies (ICSET), Kathmandu, 2012, 196-199.

Kazem, H.A., Khatib T., Sopian K. and Elmenreich W. 2014. “Performance and feasibility
assessment of a 1.4 kW rooftop grid-connected photovoltaic power system under desertic weather
conditions”. Energy and Building 82 (2014):123-129. doi.org/10.1016/j.enbuild.2014.06.048.

Kumar A.K., Sundareswaran K., and Venkateswaran P.R. 2014. “Performance study on a grid
connected 20 kWp solar photovoltaic installation in an industry in Tiruchirappalli (India)”. Energy
and Sustainable development 23 (2014):294-304. doi:10.1016/j.esd.2014.10.002.

Kumar, Shiva B, and Sudhakar K. 2015. “Performance evaluation of 10 MW grid connected solar
photovoltaic power plant in India”. Energy Report 1(2015):184-192.
doi:10.1016/j.egyr.2015.10.001.

Kumar N.M., Kumar M. R, Rejoice P.R, and Mathew M. 2017. “Performance analysis of 100 kWp
grid-connected Si-poly photovoltaic system using PVsyst simulation tool”. Energy Procedia 117
(2017):180-189. doi:10.1016/j.egypro.2017.05.121.

Lan Z, Li J. Photovoltaic technology and electricity saving strategies for fixed- velocity-measuring
system. TELKOMNIKA Indones J Electr Eng 2014;12(6):4419–26.

65
Lechner, P. and H. Schade, 2002. “Photovoltaic thin-film technology based on hydrogenated
amorphous silicon”. Progress in Photovoltaics: Research Application. 10 (2002): 85-97.
https://onlinelibrary.wiley.com/doi/abs/10.1002/pip.412/abstract.

Makrides G, Zinsser B, Norton M, Georghiou GE, Schubert M, and Werner J.H, 2007.
Performance assessment of different photovoltaic systems under identical field conditions of high
irradiance. In: Procceding of 4th Photovoltaic Science Application and Technology Conference;
Edinburgh, UK. pp. 199–202.

Martifer. 2010. “Plano de Electrificacao de Timor-Leste com Base em Energia Renováveis:

Relatório” Dili. MTR.

Maycock, P., 2011. World PV market: Technology & costsPV. Accessed on 28 November 2018.
Available at: http://www.ncsl.org/print/energy/PMaycockSolar1007.pdf.

Milosavljevic,D.D., Pavlovic,T.M., and Pirsl D.S, 2015. “Performance analysis of a grid-

connected solar PV plant in Nis, Republic of Serbia”. Renewable and Sustainable Energy Reviews
44 (2015):423-435. doi:10.1016/j.rser.2014.12.031.

Ministry of New and Renewable Energy (MNRE). 2013. “Report on developmental impacts and
sustainable governance aspects of renewable energy projects”. Accessed at 3rd of June 2018.
https://mnre.gov.in/file-manager/UserFiles/report-on-developmental-impacts-of-RE.pdf.

Mint, P., 2009. “Photovoltaic manufacturer shipment, capacity, & competitive analysis 2008/2009.
Report: NPS-Supply4. Palo Alto, California: Navigant Consulting Photovoltaic Service Program.

Nemet, G.F., 2006. “Beyond the learning curve: Factors influencing cost reduction in
photovoltaics. Energy Policy. 34 (2006): 3218-3232. doi: 10.1016/j.enpol.2005.06.020.

Nesamalar, Drusila J.J, Venkatesh S., and Raja S.C, 2017. “The drive of renewable energy in
Tamilnadu: Status, barriers and future prospect”. Renewable and Sustainable Energy Reviews 73
(2017):115-124”. doi:10.1016/j.rser.2017.01.123.

National Renewable Energy Laboratory (NREL). 2018. Available at: http://www.nrel.gov/ .

accessed on 27 November 2018.

Office of Faculty of Engineering, Science, and Technology. Universidade Nacional Timor

Lorosa’e. 2018. Dili. Timor-Leste.

Oriti, G., and Julian A.L., 2011. “Three-phase VSI with FPGA-based multi-sampled space vector
modulation, IEEE Trans Ind Application. 47 (2011): 1813-1820.

Osborne M (2010) PV Tech. GaAs solar cell from Spire sets 42.3% conversion efficiency record.
Available at: (Accessed on 27 November 2018).
https://www.pv_tech.org/news/gaas_solar_cell_from_spire_sets_42.3_conversion_efficiency_rec
ord.

66
Osborne M. (2014) PVTECH. Siva Power claims 18.8% lab CIGS efficiency. Available at:
https://www.pv-tech.org/news/siva_power_claims_18.8_lab_cigs_efficiency. (accessed 27
November 2018).

Padmavathi, K. and Daniel, S.A, 2013. “Performance analysis of a 3 MWp grid connected solar
photovoltaic power plant in India”. Energy for Sustainable Development 17 (2013):615-625.
doi:10.1016/j.esd.2013.09.002.

Parida, B., Iniyan, S., and Goic, R. 2011. “ A review of solar photovoltaic technologies”.
Renewable and Sustainable Energy Reviews. 15 (2011): 1625-1636. doi:
10.1016/j.rser.2010.11.032.

Pérez,D.M., Aguilera J., Almonacid G., Gomez P., de la Casa J., and Aguilar J.D, 2007. “Univer
Project. Agrid connected photovoltaic system of 200 kWp at Jaén University. Overview and
performance analysis”. Solar Energy Materials & Solar Cells 91 (2007):670–683.
doi:10.1016/j.solmat.2006.12.006.

Powalla M., 2006, The R&D potential of CIS thin-film solar modules. In: Proceedings of the 21st
European photovoltaic solar energy conference, Dresden, September, pp. 1789–1795

Quansah,D.A., Adaramola,M.S., Appiah,G.K., and Isaac A.E. 2017. “ Performance analysis of

different grid-connected solar photovoltaic (PV) system technologies with combined capacity of
20 kW located in humid tropical climate”. International Journal of Hydrogen Energy 41
(2017):4626-4636. doi: 10.1016/j.ijhydene.2016.10.119.

Repins I, Conteras M, Egaas B, et al. 2008 “19.9%-efficient ZnO/CdS/CuInGaSe2 solar cell with
81.2% fill actor”. Progress in Photovoltaics: Research and Application 16: 235–239.

Salmani, A., Sadeghzadeh, S., and Naseh, R.M., 2014. Optimization and Sensitivity Analysis of
a Hybrid System in Kish, Iran. International Journal of Emerging Technology and Advanced
Engineering. 4(2014):349-355. ISSN: 2250-2459. Acessed on: 17 of Dcember 2018.
https://pdfs.semanticscholar.org/7192/0ad615eebfee3812fc0cd92d7b944fc0ed58.pdf

Sampaio, P. G. V., and Gonzáles, M. O. A. 2017. “Photovoltaic solar energy: Conceptual

framework”. Renewable and Sustainable Energy Reviews. 74 (2017):590-601. doi:
10.1016/j.rser.2017.02.081.

Satyen, K.D. 1998. Recent developments in high efficiency photovoltaic cells. Renewable
Energy. 15: 467–472.

Vikrant S, and Candel S.S, 2013. “Performance analysis of a 190 kWp grid interactive solar
photovoltaic power plant in India”. Energy 55 (2013):478-485.
doi:10.1016/j.energy.2013.03.075.

67
Sharp (2011) Sharp develops solar cell with world’s highest conversion efficiency of 36.9%.
Available at: (accessed 27 November 2018 ). http://www.sharp-
world.com/corporate/news/111104.html.

Shukla A.K., Sudhakar K., and Baredar P, 2016. “Simulation and performance analysis of 110
kWp grid-connected photovoltaic system for residential building in India: A comparative analysis
of various PV technology”. Energy Report 2 (2016):82-88. doi:10.1016/j.egyr.2016.04.001.

Streetman BG and Banerjee S. 2005. Solid State Electronic Devices, 6th ed. Upper Saddle River,
NJ: Prentice Hall.

Timor-Leste Government, Office of Direção Nacional de Meteorologia e Geofisica. 2018. Dili.

Timor-Leste.

Timor-Leste Government, Office of EDTL Baucau. 2018. Baucau District, Timor-Leste.

Timor-Leste Government. Ministry of Education. 2014. “Introduction of Clean Energy by Solar

Electricity Generation System Project: Report”. Dili. Timor-Leste.

Tripathi, L., Mishra A.K, Dubey A.K, Tripathi C.B, and t Baredar P, 2016. “Renewable energy:
An overview on its contribution in current energy scenario of India”. Renewable and Sustainable
Energy Reviews 60 (2016):226-233. doi:10.1016/j.rser.2016.01.047.

Tyag V.V., Rahim Nurul A.A., and Rahim N.A. 2013. “Progress in solar PV technology:
Reasearch and achievement”. Renewable and Sustainable Energy Reviews. 20 (2013): 443-461.
Accessed on: 27 of November 2018.
https://www.researchgate.net/publication/303439083_Review_on_recent_trend_of_solar_photov
oltaic_technology.

Upadhayaya HM, Razykov TM, Tiwari A, et al. 2007. Photovoltaics fundamentals, technology
and application. In: Goswami DY and Kreith F (eds) Handbook of Energy Efficiency and
Renewable Energy. New York: CRC Press, pp. 23-1–23-63.

UNDP, United Nations Development Program, Timor-Leste. 2014. “Promoting Sustainable

Bioenergy Production from Biomass in Timor-Leste”. Accessed on the 30th of May 2018.
https://info.undp.org/docs/pdc/Documents/TLS/Biomass_ProDoc%20270514_clean%20version.
pdf.
Walker, G.R, Sernia, P.C., 2004. “Cascaded DC–DC converter connection of photovoltaic
modules”. IEEE Trans Power Electron. 19 (2004): 1130–1139.

Wu X, Keane JC, DeHart C, et al. 2001. “High-EfficiencyCTO/ZTO/CdS/CdTe Polycrystalline

Thin-Film Cells”. NERL Report: to be presented at the NCPV Program Review Meeting,
Colorado, 14–17 October 2001. https://www.nrel.gov/docs/fy02osti/31025.pdf.

68
Zhao,J. Wang, A. and Green M.A. 1998. “19.8% efficient “honeycomb” textured
multicrystalline and 24.4% monocrystalline silicon solar cells”. Applied Physics Letter. 73 (4):
1991-1993.
https://www.researchgate.net/publication/234972975_198_efficient_honeycomb_textured_multi
crystalline_and_244_monocrystalline_silicon_solar_cells.

Zhao, J. Wang, A. and Green, M. A. 2001. “24.5% efficiency PERT silicon solar cells on SHE
MCZ substrates and cell performance on other SEH CZ and FZ substrate”. Solar Energy
Materials and Solar Cells. 66 (2001): 27-36. doi: 10.1016/S0927-0248(00)00155-0.

69
ANNEX A. INPUT PARAMETERS FOR HOMMER PRO SIMULATION TOOLS

1. Average electrical load input from 2015 to 2016

2. Electrical load input for 2015

70
3. Electrical load input for 2016

4. Average solar radiation from 2015 to 2016

71
5. Solar Radiation Data for 2015

6. Solar Radiation Data for 2016

72
ANNEX B – GRAPHS OF DIURNAL LOAD AND PV GENERATION

1. Diurnal average daily load and energy generated from 2015 to 2016

2. Diurnal average daily load and energy generated for 2015

73
3. Diurnal average daily load and energy generated for 2016

74
ANNEX C – ENERGY GENERATED BY THE PV SYSTEM

1. Monthly average energy generated by the PV system from 2015 – 2016

2. Monthly average energy generated by the PV system in 2015

3. Monthly average energy generated by the PV system in 2016

75
ANNEX D –ENERGY SUPPLY TO THE FEST BUILDINGS BY THE PV SYSTEM,
ENERGY IMPORTS FROM AND ENERGY INJECTED TO THE GRID

1. Monthly energy supply by the PV system and energy imports from the grid

2. Monthly average energy injected to the grid

76
ANNEX E – PERFORMANCE PARAMETERS VALUES

1. Monthly average, from 2015 to 2016, values of Yr, Yf, and PR

2. Monthly average, from 2015 to 2016, values of the PV system losses

3. Monthly average, from 2015 to 2016, values of the capacity factor

77
4. Monthly average values of the capacity factor for 2015 and 2016

5. PV module efficiency of the average data from 2015 to 2016

78
6. Summary of performance parameter values
2015
Metrological parameters Energies Energy yields and system losses Efficiencies
Month Energy generated, E_DC, Energy generated, E_ac, Eenergy supply to grid, Yr, Yf, Ya, Ls, PR CF
Er, (kWh/m^2/day) Ta, degree C hPV hInv hSystem
kWh kWh Eg, kWh kWh/(kWp.d) kWh/(kWp.d) kWh/(kWp.d) kWh/(kWp.d)
Jan 3.91 26.27 1004.52 809.33 431.13 3.91 3.14 4.02 0.88 0.8013 0.1349 0.1160 0.8057 0.0935
Feb 4.47 25.73 1146.37 908.80 459.30 4.47 3.52 4.59 1.06 0.7885 0.1515 0.1142 0.7928 0.0905
Mar 5.20 26.20 1333.55 1019.91 468.44 5.20 3.95 5.33 1.38 0.7607 0.1700 0.1101 0.7648 0.0842
Apr 5.33 25.77 1367.26 1008.46 493.22 5.33 3.91 5.47 1.56 0.7336 0.1681 0.1062 0.7376 0.0783
May 5.50 25.14 1411.10 984.17 489.96 5.50 3.81 5.64 1.83 0.6937 0.1640 0.1004 0.6974 0.0700
Jun 5.16 25.35 1322.89 918.71 441.72 5.16 3.56 5.29 1.73 0.6907 0.1531 0.1000 0.6945 0.0694
Jul 5.72 25.19 1468.24 996.57 499.78 5.72 3.86 5.87 2.01 0.6751 0.1661 0.0977 0.6788 0.0663
Aug 6.31 24.48 1618.30 1068.16 548.05 6.31 4.14 6.47 2.33 0.6565 0.1780 0.0950 0.6601 0.0627
Sep 6.45 24.87 1653.92 1178.75 592.88 6.45 4.57 6.62 2.05 0.7089 0.1965 0.1026 0.7127 0.0731
Oct 5.85 25.76 1501.37 1000.08 485.40 5.85 3.88 6.01 2.13 0.6625 0.1667 0.0959 0.6661 0.0639
Nov 5.42 28.93 1391.44 834.15 356.19 5.42 3.23 5.57 2.33 0.5963 0.1390 0.0863 0.5995 0.0518
Dec 4.42 28.43 1134.22 752.31 410.98 4.42 2.92 4.54 1.62 0.6597 0.1254 0.0955 0.6633 0.0634
Average 5.31 26.01 1362.76 956.62 473.09 5.31 3.71 5.45 1.74 0.7023 0.1594 0.1017 0.7061 0.0723

2016
Metrological parameters Energies Energy yields and system losses Efficiencies
Month Energy generated, E_DC, Energy generated, E_ac, Eenergy to grid, Eg, Yr, Yf, Ya, Ls, PR CF
hPV hInv
Er, (kWh/m^2/day) Ta, degree C kWh kWh kWh kWh/(kWp.d) kWh/(kWp.d) kWh/(kWp.d) kWh/(kWp.d) hSystem
Jan 4.99 28.38 1280.88 839.99 455.73 4.99 3.26 5.12 1.87 0.6523 0.1400 0.0944 0.6558 0.0619
Feb 4.63 27.06 1188.59 579.06 275.81 4.63 2.24 4.75 2.51 0.4846 0.0965 0.0702 0.4872 0.0342
Mar 5.56 27.83 1425.53 1086.92 539.01 5.56 4.21 5.70 1.49 0.7583 0.1812 0.1098 0.7625 0.0837
Apr 5.90 26.87 1515.23 878.66 416.60 5.90 3.41 6.06 2.66 0.5768 0.1464 0.0835 0.5799 0.0484
May 5.50 27.10 1410.70 605.71 266.94 5.50 2.35 5.64 3.30 0.4271 0.1010 0.0618 0.4294 0.0265
Jun 5.00 26.18 1283.66 550.27 223.64 5.00 2.13 5.13 3.00 0.4264 0.0917 0.0617 0.4287 0.0265
Jul 5.33 25.73 1368.45 579.72 309.89 5.33 2.25 5.47 3.23 0.4213 0.0966 0.0610 0.4236 0.0258
Aug 6.04 25.87 1549.65 986.35 557.95 6.04 3.82 6.20 2.38 0.6331 0.1644 0.0917 0.6365 0.0583
Sep 6.02 26.94 1545.52 1153.51 610.49 6.02 4.47 6.18 1.71 0.7423 0.1923 0.1075 0.7464 0.0802
Oct 5.77 27.54 1481.70 1121.46 614.14 5.77 4.35 5.93 1.58 0.7528 0.1869 0.1090 0.7569 0.0825
Nov 5.55 28.37 1423.37 953.60 485.62 5.55 3.70 5.69 2.00 0.6663 0.1589 0.0965 0.6700 0.0646
Dec 4.36 27.59 1119.05 534.34 247.21 4.36 2.07 4.48 2.41 0.4749 0.0891 0.0688 0.4775 0.0328
Average 5.39 27.12 1382.69 822.47 416.92 5.39 3.19 5.53 2.34 0.5847 0.1371 0.0847 0.5878 0.0521

AVERAGE
Metrological parameters Energies Energy yields and system losses Efficiencies
Month Energy generated, E_DC, Energy generated, E_ac, Eenergy supply to grid, Yr, Yf, Ya, Ls, PR CF
Er, (kWh/m^2/day) Ta, degree C hPV hInv hSystem
kWh kWh Eg, kWh kWh/(kWp.d) kWh/(kWp.d) kWh/(kWp.d) kWh/(kWp.d)
Jan 4.45 27.33 1142.70 824.66 443.43 4.45 3.20 4.57 1.37 0.72 0.1374 0.1039 0.7217 0.0750
Feb 4.55 26.39 1167.48 743.93 367.55 4.55 2.88 4.67 1.79 0.63 0.1240 0.0918 0.6372 0.0585
Mar 5.38 27.02 1379.54 1053.41 503.73 5.38 4.08 5.52 1.44 0.76 0.1756 0.1100 0.7636 0.0840
Apr 5.62 26.32 1441.24 943.56 454.91 5.62 3.66 5.76 2.11 0.65 0.1573 0.0943 0.6547 0.0617
May 5.50 26.12 1410.90 794.94 378.45 5.50 3.08 5.64 2.56 0.56 0.1325 0.0811 0.5634 0.0457
Jun 5.08 25.76 1303.27 734.49 332.68 5.08 2.85 5.21 2.37 0.56 0.1224 0.0812 0.5636 0.0457
Jul 5.53 25.46 1418.34 788.14 404.84 5.53 3.05 5.67 2.62 0.55 0.1314 0.0800 0.5557 0.0445
Aug 6.17 25.18 1583.97 1027.26 553.00 6.17 3.98 6.34 2.35 0.65 0.1712 0.0934 0.6485 0.0606
Sep 6.23 25.91 1599.72 1166.13 601.69 6.23 4.52 6.40 1.88 0.73 0.1944 0.1050 0.7290 0.0765
Oct 5.81 26.65 1491.53 1060.77 549.77 5.81 4.11 5.97 1.85 0.71 0.1768 0.1024 0.7112 0.0728
Nov 5.48 28.65 1407.41 893.87 420.91 5.48 3.46 5.63 2.16 0.63 0.1490 0.0915 0.6351 0.0581
Dec 4.39 28.01 1126.64 643.33 329.10 4.39 2.49 4.51 2.01 0.57 0.1072 0.0822 0.5710 0.0470
Average 5.35 26.57 1372.73 889.54 445.00 5.35 3.45 5.49 2.04 0.64 0.1483 0.0933 0.6480 0.0605

79
ANNEX F – SIMULATION RESULTS OF THE PV SYSTEM WITH BATTERY

1. Simulation results of the PV system with UET ReFlex-100kW batteries for the average data from 2015 to 2016

80
2. Simulation results of the PV system with CELLCUBE FB20-100 batteries for the average data from 2015 to 2016

81
3. Simulation results of the PV system with UET ReFlex-100kW batteries for the average data of 2015

82
4. Simulation results of the PV system with CELLCUBE FB20-100 batteries for the average data of 2015

83
5. Simulation results of the PV system with UET ReFlex-100kW batteries for the average data of 2016

84
6. Simulation results of the PV system with CELLCUBE FB20-100 batteries for the average data of 2016

85
7. Simulation results of the sensitivity analysis of the PV system with load fluctuation

86
8. Simulation results of the sensitivity analysis of the PV system plus battery with load fluctuation

87
ANNEX G – DIURNAL AVERAGE LOAD DEMAND
1. Diurnal average load demand from 2015 to 2016

LOAD OF BUILDING (KWh)

HOUR
2015 2016 Average
1 9.37 4.41 6.89
2 9.18 4.25 6.71
3 9.05 4.16 6.60
4 9.01 4.07 6.54
5 9.00 4.11 6.56
6 9.35 4.78 7.06
7 11.61 6.65 9.13
8 25.43 16.04 20.73
9 37.48 29.01 33.25
10 51.75 42.90 47.32
11 58.47 50.47 54.47
12 60.60 53.04 56.82
13 59.05 52.07 55.56
14 57.72 50.47 54.10
15 56.21 48.30 52.25
16 48.09 40.73 44.41
17 34.95 28.56 31.76
18 18.98 13.61 16.29
19 12.11 7.21 9.66
20 11.68 5.55 8.61
21 10.91 5.19 8.05
22 10.31 5.06 7.68
23 9.87 4.83 7.35
24 9.53 4.65 7.09
TOTAL 639.69 488.31 564.91

88
2. Diurnal average load demand for 2015

MONTH OF THE YEAR 2015, kWh

Hour AVERAGE
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
0 9.79 10.44 11.74 11.56 11.43 11.25 9.91 9.45 9.07 6.28 7.57 3.90 9.37
1 9.69 10.32 11.55 11.31 11.27 11.08 9.69 9.36 8.91 5.97 7.39 3.64 9.18
2 9.62 10.26 11.45 11.14 11.14 11.01 9.60 9.25 8.81 5.75 7.11 3.49 9.05
3 9.61 10.28 11.37 11.20 11.07 11.01 9.74 9.19 8.72 5.49 6.91 3.54 9.01
4 9.63 10.32 11.44 11.15 11.10 11.01 9.80 9.14 8.70 5.31 6.79 3.57 9.00
5 9.79 10.49 11.71 11.37 11.33 11.30 10.09 9.53 9.13 6.42 7.40 3.59 9.35
6 10.73 11.72 13.80 13.63 13.04 13.04 10.97 11.19 11.71 10.82 12.20 6.40 11.61
7 23.87 25.23 30.22 27.51 25.88 24.67 21.21 23.52 28.14 28.03 25.82 21.07 25.43
8 33.42 36.77 46.29 40.23 40.92 37.49 35.73 38.27 40.63 35.97 35.44 28.62 37.48
9 44.69 46.85 62.58 57.88 54.21 51.62 49.88 53.20 57.72 51.91 51.17 39.28 51.75
10 49.29 52.99 70.80 64.50 57.03 56.84 58.53 60.91 66.18 60.76 59.72 44.06 58.47
11 48.26 57.61 68.18 63.38 56.48 59.67 62.19 64.54 72.71 66.09 61.55 46.50 60.60
12 44.15 55.33 63.71 64.02 57.19 59.22 61.13 64.12 72.11 64.35 59.73 43.57 59.05
13 43.53 53.78 65.87 65.65 61.61 57.18 59.70 61.07 69.37 61.36 55.94 37.61 57.72
14 40.28 51.66 63.81 56.06 60.33 58.89 60.71 61.77 68.96 59.63 55.15 37.27 56.21
15 33.63 43.15 54.61 49.47 52.92 50.41 54.04 53.38 60.00 49.79 44.75 30.96 48.09
16 23.53 32.64 40.95 35.53 39.61 38.04 40.63 38.87 44.26 32.77 30.98 21.55 34.95
17 16.77 19.38 20.64 19.72 18.94 20.54 21.30 21.83 21.40 16.75 17.28 13.19 18.98
18 10.71 12.32 12.71 13.60 12.97 13.76 12.73 13.62 13.01 10.25 11.53 8.14 12.11
19 11.08 12.03 13.90 14.35 13.90 13.68 12.03 12.49 11.52 8.60 10.25 6.31 11.68
20 11.06 12.01 13.71 13.56 13.40 13.33 11.44 11.34 10.23 7.29 8.83 4.76 10.91
21 10.63 11.65 12.86 13.24 12.67 12.40 10.83 10.55 9.56 7.00 8.03 4.27 10.31
22 10.34 11.09 12.28 12.49 12.13 11.78 10.30 10.03 9.27 6.79 7.86 4.03 9.87
23 9.95 10.67 11.95 11.97 11.81 11.10 10.09 9.73 9.14 6.53 7.73 3.69 9.53

TOTAL 534.05 618.97 748.13 704.53 682.38 667.12 662.27 676.38 728.82 619.90 607.10 413.71 639.69

89
3. Diurnal average load demand for 2016

MONTH OF THE YEAR 2016, KWh

HOUR
JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER AVERAGE
1 3.12 4.90 5.21 4.81 5.13 5.24 4.21 2.24 4.24 2.67 7.29 3.89 4.41
2 2.98 4.69 5.00 4.47 5.03 5.28 3.99 2.05 3.97 2.43 7.19 3.89 4.25
3 2.85 4.91 4.91 4.19 4.74 5.01 3.80 1.97 3.79 2.26 7.20 4.23 4.16
4 2.72 4.71 4.66 3.82 4.62 5.03 3.75 1.73 3.74 2.58 7.12 4.33 4.07
5 2.80 5.17 4.98 4.32 4.78 4.94 3.66 1.47 3.56 2.42 7.01 4.22 4.11
6 2.77 6.26 6.02 5.57 5.16 5.80 3.46 2.40 4.43 3.62 7.34 4.48 4.78
7 3.72 6.75 6.59 5.54 5.99 7.84 5.24 5.49 7.06 8.42 10.73 6.39 6.65
8 18.97 18.48 22.92 17.75 11.37 10.20 8.93 11.21 16.95 20.60 21.20 13.90 16.04
9 29.82 27.86 38.93 34.39 23.96 19.17 16.55 24.40 36.40 39.57 35.18 21.94 29.01
10 42.44 39.82 55.83 50.44 37.54 31.93 26.98 41.42 51.72 54.80 51.55 30.33 42.90
11 48.79 43.49 65.91 59.42 44.65 40.52 33.96 51.11 59.45 62.46 59.18 36.75 50.47
12 50.98 46.44 68.47 59.36 47.17 42.77 35.10 55.57 66.79 65.11 59.94 38.73 53.04
13 47.75 46.98 66.72 56.69 44.72 44.41 34.58 54.98 68.82 64.26 58.08 36.86 52.07
14 44.25 45.73 63.92 55.39 43.64 43.65 35.10 53.71 68.66 60.50 55.90 35.24 50.47
15 42.53 44.42 66.43 54.57 40.33 42.25 33.34 52.46 66.85 54.48 49.72 32.21 48.30
16 35.41 38.19 54.40 44.61 35.44 36.28 28.61 45.65 57.00 46.19 41.78 25.26 40.73
17 23.10 27.70 39.13 32.29 25.48 25.82 19.97 30.75 39.60 31.50 30.01 17.38 28.56
18 14.06 13.96 21.38 15.90 11.38 11.50 9.88 13.25 15.27 12.51 14.10 10.11 13.61
19 6.50 8.18 9.55 8.50 7.41 7.47 5.41 5.63 6.80 5.80 9.29 5.93 7.21
20 4.52 5.80 7.05 6.66 6.21 6.33 4.36 3.51 5.37 4.45 8.01 4.32 5.55
21 4.00 5.58 6.50 6.16 5.79 6.01 4.17 3.00 4.92 4.34 7.87 3.97 5.19
22 3.89 5.22 6.13 5.97 5.66 5.66 4.46 2.96 4.93 3.99 7.77 4.08 5.06
23 3.88 5.07 5.71 5.49 5.47 5.77 4.29 2.46 4.64 3.68 7.57 3.96 4.83
24 3.56 4.89 5.56 5.11 5.31 5.60 4.26 2.40 4.46 3.32 7.52 3.76 4.65
TOTAL 445.40 447.77 641.91 551.39 437.00 424.47 333.81 471.82 609.41 561.97 578.56 356.14 488.31

90
4. Diurnal average energy generated by the System

ENERGY GENERATED (kWh)

HOUR
2015 2016 Average
1 0.00 0.00 0.00
2 0.00 0.00 0.00
3 0.00 0.00 0.00
4 0.00 0.00 0.00
5 0.00 0.00 0.00
6 0.00 0.00 0.00
7 0.42 0.68 0.55
8 15.68 15.40 15.54
9 57.53 50.68 54.11
10 97.78 83.78 90.78
11 121.96 106.88 114.42
12 133.75 116.90 125.33
13 138.28 118.49 128.38
14 132.97 111.23 122.10
15 112.81 95.41 104.11
16 84.75 70.46 77.60
17 48.24 39.80 44.02
18 15.39 12.79 14.09
19 1.11 0.92 1.01
20 0.00 0.00 0.00
21 0.00 0.00 0.00
22 0.00 0.00 0.00
23 0.00 0.00 0.00
24 0.00 0.00 0.00
TOTAL 958.17 850.13 892.04

91
5. Diurnal average energy generated by the system for 2015

MONTH OF THE YEAR, kWh

HOUR AVERAGE
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
6 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
7 0.29 0.04 0.03 0.04 0.02 0.00 0.00 0.01 0.28 1.55 1.67 1.10 0.42
8 17.60 14.57 16.22 13.84 10.51 6.78 5.29 10.87 22.62 27.13 21.70 21.05 15.68
9 54.87 55.57 66.50 61.53 58.66 46.80 46.36 56.86 72.40 63.52 53.51 53.79 57.53
10 86.97 84.63 111.03 108.74 101.22 92.02 93.45 104.77 117.41 102.66 87.01 83.47 97.78
11 103.84 110.31 138.75 135.23 122.02 114.03 122.25 133.50 144.59 131.06 108.33 99.61 121.96
12 107.66 130.43 142.96 142.65 129.81 129.11 135.35 148.98 161.92 143.05 120.48 112.62 133.75
13 110.19 138.54 133.73 149.63 137.32 135.96 145.62 157.47 168.90 146.73 123.66 111.61 138.28
14 116.26 129.17 134.00 145.89 139.08 128.96 143.94 149.01 160.88 135.90 117.16 95.46 132.97
15 96.26 106.47 119.60 103.04 121.27 116.23 126.93 128.01 141.21 115.71 94.76 84.19 112.81
16 69.82 71.96 87.28 83.14 97.71 89.68 101.62 100.77 109.65 80.55 63.86 60.96 84.75
17 35.66 45.34 53.49 47.16 58.07 50.63 59.39 59.51 62.90 41.07 32.63 33.03 48.24
18 18.73 22.70 18.85 16.14 12.80 12.88 16.03 17.94 15.64 11.01 9.16 12.76 15.39
19 3.25 3.69 1.78 1.45 0.15 0.15 0.34 0.45 0.34 0.14 0.20 1.33 1.11
20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
21 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
TOTAL 821.39 913.42 1024.23 1008.46 988.65 918.71 996.57 1068.16 1178.75 1000.08 834.15 752.31 958.74

92
6. Diurnal average energy generated by the system for 2016

MONTH OF THE YEAR, kWh

HOUR AVERAGE
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
6 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
7 0.13 0.05 0.05 0.04 0.02 0.00 0.00 0.02 0.40 2.41 3.75 1.29 0.68
8 15.99 14.90 16.80 12.16 6.19 4.03 3.48 10.55 23.06 35.42 31.43 16.09 15.84
9 50.19 60.07 65.69 56.54 31.33 24.45 27.16 50.85 73.80 83.99 69.47 36.07 52.47
10 84.60 93.18 102.52 96.67 62.52 50.91 57.16 96.91 113.14 123.72 104.25 53.09 86.55
11 109.30 112.06 136.75 121.50 82.12 71.35 78.24 124.36 136.05 149.08 130.95 70.82 110.21
12 119.05 131.12 151.35 123.94 90.27 76.33 86.33 140.72 159.23 160.99 135.04 78.08 121.04
13 117.52 129.41 155.04 124.16 87.88 84.63 85.21 144.53 166.92 162.15 133.53 76.72 122.31
14 112.43 123.62 147.78 117.98 83.77 78.82 81.64 135.88 157.66 143.92 123.70 71.75 114.91
15 98.35 101.21 134.78 97.56 69.54 70.35 73.95 121.86 139.32 114.82 99.26 60.06 98.42
16 75.54 72.45 96.81 70.01 53.53 52.41 56.75 94.51 103.57 83.47 71.33 38.44 72.40
17 39.45 40.52 55.20 42.63 30.50 28.86 33.31 51.49 60.87 48.17 38.77 20.82 40.88
18 15.40 14.53 21.47 14.54 7.91 7.94 9.98 14.35 18.80 13.00 11.51 9.72 13.26
19 2.02 2.64 2.69 0.92 0.13 0.20 0.30 0.33 0.68 0.32 0.61 1.38 1.02
20 0.00 1.48 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.12
21 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
TOTAL 839.99 897.22 1086.92 878.66 605.71 550.26 593.51 986.35 1153.51 1121.46 953.60 534.34 850.13

93
ANNEX H – PV SYSTEM CONFIGURATION AND MANAGEMENT SYSTEM

1. PV system configuration

94
2. Diagram of PV system data management and control system

95