IN
DEGREE PROJECT
ENERGY AND ENVIRONMENT,
SECOND CYCLE, 30 CREDITS
,
STOCKHOLM SWEDEN 2019
On the Profitability of
Large-scale PV Plants in Sweden
Site Selection, Grid Connection and Design
ANNELIE WESTÉN
KTH ROYAL INSTITUTE OF TECHNOLOGY
KTH Royal Institute of Technology
School of Electrical Engineering and Computer Science
Degree project in Sustainable Energy Engineering
On the Profitability of Large-scale PV Plants in Sweden: Site Selection,
Grid Connection and Design
Author: Annelie West´en
Supervisor: Abolfazl Khodadadi, KTH Royal Institute of Technology Supervisor:
Examiner:
Mikael Ronge, Eneo Solutions
Abstract
The market for large scale PV plants in Sweden is growing, with six PV plants of 1 MWp or more being installed today. The size of the newly installed plants has increased from 1 MWp to 5.5 MWp during the last 5 years. As the market and size of the plants continue to grow, larger investments, risks and possible profits will be built into the project. The site selection will affect the project in terms of power production, grid connection, plant design, land lease, among many things. This report focuses on how these early choices in the development of the project affect the profitability of the PV plant.
A literature study has been conducted for in-depth knowledge of PV plants as well as the most important aspects of a pre-feasibility study. The literature study has a specific focus on the components and design of a PV plant, the grid connection, economy and future changes of the electricity price in Sweden. One important conclusion from the literature study is that the company which develops a new PV plant should contact the grid company in the area of interest as soon as possible, due to the time-consuming process of getting an approval for the new connection as well as that a major reinforcement of the grid may ruin the prospect of the PV plant project. Another conclusion is that the site should be selected to maximize the solar irradiation, meanwhile minimizing the costs for upgrading or reinforcing the power grid. The economic case of a PV plant project can generally be improved by choosing a site in the southern part of Sweden. Benefits are gained from both higher solar irradiation compared to the northern part of Sweden as well as compensations from the grid company for reducing losses in the power grid. The land should preferably have a dual purpose, meaning that the same land can be used for both the PV plant and another purpose such as a wind farm, airport, landfill or feeding small lively stocks of sheep, etc.
A case study has been conducted at three sites in the southern part of Sweden, located on ¨Oland, Skurup, and Stenungsund. The sites are compared by evaluating the prospect of two different sizes of PV plants. The PV plant performance is evaluated by a model developed in the program MATLAB. Hourly data of the solar irradiation is gatherer for the three sites for 2017 and 2018, using the system STR˚ANG which is given by the Swedish Meteorological and Hydrological Institute (SMHI). Hourly data of the temperature is gathered from the closest weather station to the site, also from SMHI. The sites are compared by power production and resulting LCOE values, including specific grid tariffs, land leasing costs and typical investment costs for PV plants. The investment costs are gathered from recent reports, cross-checked with costs for the largest PV plants built in Sweden.
Sammanfattning
Marknaden f¨or storskalig solkraft v¨axer i Sverige. Idag finns det 6 stycken solparker av storleken 1 MWp eller st¨orre. Storleken av de nya solparkerna har v¨axt fr˚an 1 MWp to 5.5 MWp under de senaste 5 ˚aren. Med en v¨axande marknad och storlek p˚a solparkerna, v¨axer ¨aven investeringskostnaden, risker och m¨ojliga f¨ordelar i projekten. Platsvalet f¨or solparken p˚averkar produktionen, n¨atanslutningen, designen av parken samt markkostnader. Denna rapport fokuserar p˚a hur dessa tidiga val i projektet p˚averkar projektets l¨onsamhet. En litteraturstudie har genomf¨orts f¨or att ¨oka f¨orst˚aelsen av en solpark och de viktigaste delarna i en l¨onsamhetsanalys. Litteraturstudien fokuserar p˚a komponenter i en solpark, n¨atanslutning, platsval, ekonomi och framtida f¨or¨andringar av elpriset i Sverige. En viktig slutsats fr˚an litteraturstudien ¨ar att ett f¨oretag som utvecklar en ny solpark b¨or kontakta n¨atbolaget s˚a tidigt som m¨ojligt. Processen f¨or att f˚a en godk¨and nyanslutning till n¨atet kan vara tidskr¨avande samt att en omfattande f¨orst¨arkning av n¨atet kan vara s˚a dyrt att projektet inte l¨angre blir aktuellt att genomf¨ora. En annan slutsats ¨ar att platsvalet b¨or fokusera p˚a att maximera solinstr˚alningen och minimera kostnader f¨or kraftn¨atet och infrastruktur. L¨onsamheten f¨or solparksprojektet kan generellt s¨att ¨okas genom att v¨alja en plats i s¨odra Sverige, b˚ade f¨or den h¨ogre solinstr˚alningen j¨amf¨ort med norra Sverige samt att det ger f¨ordelar med n¨atkompensationer f¨or minskade f¨orluster i n¨atet. Marken kan med f¨ordel anv¨andas med ett dubbelt syfte, vilket syftar till att den anv¨ands b˚ade f¨or solparken samt ett ytterligare syfte s˚a som vindkraft, en flygplats, deponier, eller som betesmark f¨or f˚ar eller liknande.
Tre potentiella platser f¨or solparker har analyserats i s¨odra Sverige. Specifik mark ¨ar vald p˚a ¨Oland, i Skurup och i Stenungsund. Produktionen fr˚an tv˚a olika storlekar av solparker har utv¨arderats p˚a de tre platserna med hj¨alp av en modell som utvecklats i MATLAB. Timdata f¨or solinstr˚alning fr˚an 2017 och 2018 har h¨amtats fr˚an systemet STR˚ANG som ¨ags av SMHI. Timdata f¨or temperaturen har h¨amtats fr˚an den n¨armsta v¨aderstation till respektive plats, ¨aven denna data kommer fr˚an SMHI. Platserna j¨amf¨ors med avseende p˚a kraftproduktionen fr˚an solparkerna samt det resulterande LCOE v¨ardet. LCOE v¨ardet inkluderar specifika n¨attariffer, markkostnader och investeringskostnader f¨or solparker fr˚an rapporter som ¨
aven j¨amf¨orts med kostnaderna f¨or de tv˚a st¨orsta solparkerna i Sverige.
N˚agra slutsatser fr˚an j¨amf¨orelsen mellan de tre platserna ¨ar att solinstr˚alningen har en stor betydelse f¨or l¨onsamheten f¨or en solpark, dock l¨agre ¨an ett 1:1 samband mellan ¨okad solinstr˚alning och ¨okad produktion fr˚an solparken. En optimering av ett antal designparametrar f¨or solparkerna resulterade i liknande v¨arden f¨or de tre platserna, med undantag f¨or avst˚andet mellan raderna av moduler som beror p˚a den lokala latituden. En unders¨okning har gjorts av hur skiftande elpriser kan p˚averka framtida solparker. Fluktuerande elpriser under ett dygn med l˚aga eller noll-pris perioder mitt p˚a dagen implementerades f¨or timproduktionen fr˚an olika solparker. Detta anv¨andes f¨or att unders¨oka fr˚agan om ifall det skulle vara b¨attre att rikta solcellerna ¨
Contents
List of Figures 7 List of Tables 8 1 Introduction 9 1.1 Background . . . 10 1.2 Purpose . . . 10 1.3 Delimitations . . . 11 1.4 Research methodology . . . 11 2 Frame of reference 12 2.1 PV plants . . . 12 2.1.1 Components in a PV plant . . . 132.1.2 Design of PV plant layout . . . 18
2.1.3 Performance . . . 18
2.1.4 Operation and maintenance . . . 19
2.2 Grid connection . . . 19
2.2.1 The Swedish power grid . . . 19
2.2.2 Regulations . . . 21
2.2.3 Grid tariffs . . . 22
2.2.4 Connecting a PV plant to the grid . . . 23
2.2.5 Physical constraints in the grid . . . 25
2.2.6 Grid companies in Sweden . . . 26
2.3 Siting . . . 26 2.4 Economics . . . 29 2.4.1 Economic model . . . 29 2.4.2 Expected prices . . . 30 2.5 Future scenarios . . . 32 3 Model 33 3.1 PV Plant Layout . . . 33 3.1.1 Initial design . . . 34
3.1.2 Layout and area . . . 34
3.2 PV Plant Performance . . . 35
3.2.1 Input parameters . . . 36
3.2.2 PV modules . . . 38
3.2.3 Hourly power output . . . 39
3.2.4 Losses . . . 39
3.3 PV Plant Optimization . . . 40
3.4 Verification of the model . . . 40
3.5 Uncertainties . . . 41
4 Case study 42 4.1 Baseline systems . . . 42
4.2 Siting . . . 42
4.2.1 Methodology of siting . . . 42
4.2.2 Sites for further investigation . . . 43
4.2.3 Input data for the three sites . . . 45
4.2.4 Grid tariffs . . . 46
5 Results and discussion 50
5.1 Plant performance . . . 50
5.2 Future scenarios . . . 51
5.2.1 Scenario 1 . . . 52
5.2.2 Scenario 2 . . . 54
5.3 Comparison of the three sites . . . 56
5.4 Sensitivity analysis . . . 57
6 Conclusions 60
References 61
Definitions
• PV - Short for Photovoltaic, which is a technique for capturing solar energy.
• PV plant - A system of PV modules (and other essential components), installed to produce electricity. It is generally referring to a plant with a capacity larger than 1 MWp in this report, if nothing else is stated.
• Wp - ”Watt-peak”, the SI-unit for power combined with a ”peak” which represents that it is the installed peak value.
• Wh - ”Watt-hour”, the unit for energy which is the power multiplyed with time. • I - Current, measured in Ampere.
• V - Voltage, measured in Volt.
• Vth - Thermal equivalent of voltage in the PV module, measured in Volt
• Iph - Photocurrent, measured in Ampere.
• Isat- Saturation current of the PV module, measured in Ampere.
• Rs - The series resistance in a PV module, measured in Ohm.
• Rp - The parallel resistance (sometimes called shunt resistance) in a PV module, measured in Ohm.
• A - The diode ideality factor.
• Ns- The number of PV cells connected in series in a PV module.
• MPP - Maximum Power Point, the operating point of the module which results in the maximum power being produced, measured in Watt.
• MPPT - Maximum Power Point Tracker, is the device that makes the operate at (or close to) the MPP.
• STC - Standard Test Conditions, test conditions commonly used in the industry. The values are: Irradiation = 1000 W/m2, module temperature = 25◦C and an air mass = 1.5.
• NOCT - Nominal Operating Cell Temperature, test conditions commonly used in the industry. The values are: Irradiation = 800 W/m2, ambient temperature = 20◦C and and a wind velocity = 1 m/s.
• Inverter Load Ratio (ILR) - The ratio between the installed capacity of the PV modules connected to the inverter and the rated AC power of the inverter.
• Grid code - Rules for the design, operation, maintenance, and development of the grid.
• PPA - ”Power Purchase Agreement”, a specific form of selling energy from a power plant to a specific predetermined price during a specific period.
• LCOE - ”Levelized Cost Of Energy”, represents the cost for the converted energy from a specific power plant. All costs during the lifetime of the plant are divided by all the electricity projected to be sold during its lifetime.
List of Figures
1.1 An overview of the existing large-scale PV plants in Sweden. . . 9
2.1 A PV plant in Varberg, originally published on Solkompaniet.se. . . 12
2.2 A picture of PV modules, published with permission from Eneo Solutions/Shutterstock. . . 13
2.3 An illustration of a IV-curve for PV modules. . . 14
2.4 The impact of ILR on the output power. . . 15
2.5 An example of mounting systems. . . 16
2.6 An illustration of an east-west mounting system seen from the side. . . 16
2.7 A rooftop PV plant in Sweden during winter. . . 17
2.8 The four grid areas in Sweden. . . 20
2.9 The process for connecting a new plant to the grid. . . 24
2.10 What a substation in the transmission grid may look like, published with permission. . . 25
2.11 Global solar irradiation map for Sweden. . . 27
2.12 The two figures illustrates population density and installed PV capacity in Sweden. . . 28
2.13 A site in Germany with both wind power and solar power operating in a close distance. . . 29
3.1 An overview a PV plant layout. . . 33
3.2 An illustration of the distance between rows of PV modules. . . 35
3.3 An overview of the solar angles used in the model. . . 37
3.4 The figures illustrates the behaviour of a PV module in the model. . . 40
3.5 Model validation. . . 41
4.1 An overview of the location of the three sites. . . 44
4.2 Historical global irradiation in Sweden, presented on SMHI.se. . . 45
4.3 Optimum elevation angles. . . 47
4.4 Optimum inverter load ratios (ILR). . . 48
5.1 Monthly power production during 2018. . . 50
5.2 Power production from different PV plant orientations and a fluctuating electricity price. . . 52
5.3 Assumed variations and prices for electricity at the spot market. . . 53
5.4 The results from implementing the price curves. . . 54
5.5 Possibilities for a profitable PV plant on the spot market for electricity. . . 55
5.6 Sensitivity analysis of irradiation. . . 58
5.7 Sensitivity analysis of module temperature. . . 58
5.8 Sensitivity analysis of wind speed. . . 59
B.1 Optimum orientation. . . 67
List of Tables
2.1 Typical losses in a PV plant [12]. . . 18
3.1 Design choices for the PV plant. . . 34
3.2 Values of coefficients used in the equation of time. . . 38
4.1 Initail design choices for the two baseline PV plant. . . 42
4.2 The two baseline PV plants, facing south. . . 42
4.3 Exported electricity from the four grid areas during 2018. . . 43
4.4 Tariffs for feeding electricity to the grid. . . 46
4.5 Calculated distance between rows for two reference days. . . 47
4.6 Values from the graphs above. . . 48
4.7 The final PV plants, optimized from the 4 MWp baseline PV plant. . . 49
4.8 The final PV plants, optimized from the 10 MWp baseline PV plant. . . 49
5.1 PV plant performance. . . 50
5.2 Values used in the economic analysis. . . 51
5.3 Economic analysis. . . 51
5.4 Irradiation data for the sites . . . 56
5.5 Total irradiation on the module surface for the three sites, 2017-2018. . . 56
5.6 Final performance from the case study. . . 56
5.7 Plant performance of the baseline PV plants. . . 57
5.8 A comparison of irradiation and power production between the sites. . . 57
5.9 Induced losses due to temperature. . . 59
A.1 Global irradiation at the three sites during 2018, presented on a monthly basis. . . 65
A.2 Comparing diffuse and direct irradiation during 2018. . . 65
A.3 Representative values for the temperature at the three sites during 2018. . . 65
A.4 Global irradiation at the three sites during 2017, presented on a monthly basis. . . 66
A.5 Comparing diffuse and direct irradiation during 2017. . . 66
A.6 Representative values for the temperature at the three sites during 2017. . . 66
B.1 Values for the optimum inverter load ratios and the saved money. . . 68
B.2 The final PV plants with an east-west orientation. . . 68
1
Introduction
The market for grid-connected PV plants in Sweden is growing. The total capacity of solar power connected to the Swedish power grid increased with 65% from 2016 to 2017, an increase from 140 MW to 231 MW [1]. The first PV plant with an installed capacity of 1 MW or more, was installed outside of V¨aster˚as during 2014. The market for large scale PV plants has increased with approximately one new plant per year. During Q1 2019, the sixth and largest PV plant were installed in G¨oteborg, with the installed capacity of 5.5 MW [2].
The six PV plants with an installed capacity of 1 MW or higher differ in terms of design and business case. Two of the PV plants are mounted on rooftops, one PV plant has a dual-axis tracking system and three PV plants are ground mounted with a fixed angle. A summary of the PV plants is shown in the following illustration, with the placement of the plants, installed capacity, the year of commissioning as well as the type of plant [2, 3, 4].
Figure 1.1: An overview of the existing large-scale PV plants in Sweden.
The PV plants have different business cases, the two rooftop plants are owned by companies that use their solar power production in the building and therefore reducing their electricity cost. The PV plant in S¨ave uses a business case of letting persons invest in small parts of the plant, sometimes owning only one module in the plant.
100 MW. PPAs are seen in Sweden for medium-sized PV plants up till today, a majority of these are signed with the company Eneo Solutions [5, 6].
A trend can be noted from the six large scale PV plants in Sweden, namely that the installed capacity of new plants increases with time. As the capacity of the PV plants increases, so does the risks, possible environmental impact as well as the profit of the project. Higher investments in a project often lead to higher requirements stated by the investors. The economic aspect of a large PV plant project is an important factor. Choosing the right location for the project may be vital for the project economy, due to factors such as the local solar irradiation, grid connection and finding available land that is large enough.
1.1
Background
This report presents a master thesis project conducted during the final semester of a technical education at KTH Royal Institute of Technology. The master thesis is a 30 credits course, conducted at the School of Electrical Engineering and Computer Science at the Department of Electric Power and Energy Systems. The project is carried out for the company Eneo Solutions (Eneo). The company was founded in 2013 and has 25 employees today. The stated purpose of Eneo is:
”Leading the way in the change towards more renewable power production”
Some of the ways in which the purpose is turned into actions are by sharing knowledge, helping large organizations to deploy solar power with its full value for the environment and the organization as well as always taking the long-term perspective. Eneo has pioneered corporate PPAs for solar power on the European market and built more than 45 PV plants and geothermal plants during the first five years of the company’s existence.
Eneo offers both PPAs and leasing solutions for solar power to corporate and public customers. This includes all aspects of the project, some of which are: strategy specific for each organization, development, procurement, construction, financing of the PV plant as well as Operation and maintenance services. The same can be deployed as consulting services as well if the organization prefers to own the plant themselves.
1.2
Purpose
The main purpose of this project is to contribute to the development of the Swedish PV plant market. There are many aspects of interest. This project has embraced the perspective of project companies developing PV plants. The chosen focus areas are :
• Increasing knowledge about site selection
• Investigating the process of connecting a new PV plant to the grid at different locations in Sweden • Investigating if the optimization of a PV plant design differs with the chosen site
• Investigating how the design and prospects of a PV plant would be affected by different possible changes in the future.
1.3
Delimitations
The master thesis project is limited to 100 days, which results in the requirement of limiting the project. The project is limited to locations in Sweden, 3 locations will be chosen for simulating the performance of a grid-connected PV plant. With only a few sites and therefore only a small fraction of the possible grid companies in Sweden being investigated, the results are not general for all possible sites in Sweden since it is based on a few specific cases.
The economic model includes the main aspects of the project. Since it is often difficult to receive real data of components, the material cost will rather be included as statistical costs per installed kWp found in literature.
The components used in the PV plant will be chosen from industry custom and commonly used brands. No specific optimization will be conducted for the choice of the type or brand of the components. The panels will be polycrystalline, no tracking system and no batteries will be considered. The project only considers power generation from ground-mounted PV plants that are connected to the grid. Furthermore, there are only a few of all the phases of a PV plant project that are included in this report. The phases being investigated are site selection, grid connection, and cost optimization.
1.4
Research methodology
2
Frame of reference
The purpose of this chapter is to present important theory for the upcoming chapters in the report. It consists of the five parts: PV plants, Grid connection, Siting, Economy and Future scenarios.
The first part of the chapter presents an overview of PV plants, with its most important components and their purpose. The performance of a PV plant will then be presented followed by a design perspective of the plant layout and finally O&M.
Regarding the second part, presenting the perspective of Grid connection, the information goes further than solely the process and the physical connection of the PV plant to the grid. The idea is to give a project company which develops PV plants a better understanding of statements, regulations and tariffs from a grid company as well as increasing their knowledge about their rights during the collaboration with a grid company.
The last three parts of Siting, Economy and Future scenarios takes on a perspective of the most important information for enabling a pre-feasibility study.
2.1
PV plants
The sight of a PV plant may strike the eye as only being hundreds of PV modules placed in neat rows. However, the modules accompanied by the other technical components can be coupled in many different ways which highly affects the power production and plant performance. All of the components have a purpose that is fully, partly or not fulfilled due to the many design aspects of type, size, and quality of the components. The main components, common designs, and PV plant performance will be explained in this subchapter.
2.1.1 Components in a PV plant
Before diving into a detailed level of the different components, a short overview of a general PV plant will be presented. A PV plant consists of modules connected in series and parallel, forming strings and arrays which are connected to an inverter and later to the grid. The modules produce DC power which is converted to AC power in the inverter before it is distributed to the grid. All large PV plants have transformers to elevate the voltage from the PV modules to the grid voltage.
PV modules
A PV module uses the photoelectric effect for converting sunlight to electricity. The photons in the incoming solar irradiation with higher or equal energy as the band gap in the PV cells excite free electrons in the p-n junction in the photovoltaic cell (PV cell) and induces a current when a load or inverter is coupled to the circuit. The current is directly dependant on the number of photons in the incoming insolation. Since the measured value of solar irradiation on a PV module is directly correlated with the number of photons reaching it, it can be established that the power production from a PV module is directly dependant on the irradiation. Higher irradiation results in higher current.
Figure 2.2: A picture of PV modules, published with permission from Eneo Solutions/Shutterstock. In Figure 2.2, one can distinguish the different cells in the PV modules, seen as the blue squares.
The power production from a PV module is mainly affected by the solar irradiation and the temperature of the module. In a slightly simplified manner, one can say that the irradiation affects the operating current as well as the short circuit current and that the module temperature affects the operating voltage and open circuit voltage. The short circuit current (Isc) is measured when the voltage is zero, which gives the highest
current that can occur in the system. The open circuit voltage (Voc) is measured when there is no load
The values for the operating current and voltage are correlated and can best be understood by the IV-curve of the module. A general IV curve can be seen in Figure 2.3:
Figure 2.3: An illustration of a IV-curve for PV modules.
The point on the IV-curve which results in the highest power output is called the maximum power point (MPP). It is desirable to always have the PV modules operating at the MPP, however, one should keep in mind that the MPP is constantly changing as the outer conditions change.
When PV panels are connected to form a system, a steady state production can be used for setting up an equivalent electrical model. This presumption results in the behavior of modules in series (forming a string) are adding voltage by each module and connecting strings in parallel add the current from each of the strings [8].
Power converter
The power converter does more than just convert the DC to AC power, it synchronizes the PV plant with the grid, enables the PV plant to fulfill the grid codes as well as contains the maximum power point tracker (MPPT). The MPPT optimizes the power production by making sure that the PV modules are operating at their MPP, or as close as possible.
For large scale PV plants connected to a three-phase grid, the power converter can be either a centralized inverter or an inverter with a multi-string configuration, both of which have their advantages. Centralized inverters have gained their popularity due to the simple structure, only including one MPPT, one control unit and one LF transformer. However, the simplicity and only including one of each component also means that the optimal MPP is common for all PV modules connected to the inverter. If a multi-string configuration is used instead, multiple inverters are installed in parallel which enables different MPPs in the plant which increases the efficiency.
Large PV plants often require several inverter units, due to component restraints of the inverter and the modules. Most commercialized PV modules have a maximum insulation capability of 1000 or 1500 V. The maximum capacity, current and voltage of the inverter differ with type, brand, and model.
When choosing an inverter, the DC power input stated by the inverter should range between 80-120% of the peak power from the connected PV modules. The ratio between the rated capacity from the PV modules connected to an inverter and the rated inverter capacity (AC side) is called the Inverter Load Ratio (ILV). The ILR is normally in the range of 1.15 to 1.2. The optimum value is a trade-off between inverter cost and energy loss due to the clipping of produced power [9, 10]. Clipping occurs when the rated capacity of the inverter is lower than the produced power. The concept of clipping and the inverter load ratio is illustrated in Figure 2.4:
(a) Low ILR, no clipping. (b) Increased ILR, clipping occurs.
Figure 2.4: The impact of ILR on the output power.
Figure 2.4 illustrates a low ILR to the left, where no clipping occurs and a higher ILR to the right where the orange area represents lost energy.
Another design feature is to make sure that the range of the MPPT in the inverter covers the MPPs for the modules at 1000 W/m2 in the temperature range of -10 to 70◦C [9].
An inverter has a linear dependency between efficiency and the PV plant yield. The efficiency depends on the DC voltage and operating temperature. Some inverters have a large MPP window, however, the efficiency differs along with the MPP voltage range. The designer of a PV plant should make sure that the chosen inverter has a good performance for the MPP range expected to occur most often. Both central inverters and multi-string inverters usually have a power conversion efficiency of 97%. A dual-central inverter has slightly higher efficiency, around 98% [7].
Control units
Several design features such as inverter topology as well as grid codes determine the control strategy for the PV plant. Since components and strategies for controlling the PV plant is not the main focus of this report, some features will be mentioned shortly only for the reader to get an overview.
As concluded earlier, the MPP depends both on irradiation and temperature for one PV module. When modules are included in a system that does not have an inverter for each module, the system should operate at the system MPP. Preferably it would have been calculated from measured values of the temperature and irradiation, however, a pyranometer is very expensive and rarely used for that purpose. The MPPT uses a numerical method to momentarily finding the MPP. The methods used in practice can reach up to 99% of the true MPP value.
Another control feature provided by the inverter is grid monitoring and an interaction unit to enable the PV plant to disconnect from the grid when necessary. Monitoring the grid is important for the grid synchro-nization.
Ground mounting system
A support structure is needed for tilting the PV panels, also often for elevating the panels from the ground. The structure should be sized for the module weight and local conditions such as wind and soil/snow conditions. The suitable material of the structure and the foundation depends on the type of soil at the site [9].
There are many different designs available for the mounting system. The panels can either be mounted in a standing position (the shorter side of the module towards the ground) or a laying position (the longer side of the module towards the ground). There are also numerous possibilities of the number of modules stacked beside each other in the two directions, illustrated as the black arrows in Figure 2.5:
Figure 2.5: An example of mounting systems.
Another choice is the tilt of the mounting system, it could have a fixed tilt, seasonal tilts or one- or two axis tracking. The mounting system would also differ if the orientation is east-west instead of due south. This type of system typically has a lower elevation angle and no required distance between the rows which means that more PV modules can be installed per hectare compared to orienting the modules due south. An illustration of an east-west mounting system can be seen in Figure 2.6.
Figure 2.6: An illustration of an east-west mounting system seen from the side.
Other components
One important component that is dimensioned based on the grid in which the PV plant is connected to is transformers. A common solution when the plant is connected to a grid with a voltage higher than 20 kV, is to have an LF transformer in connection to the inverter and another transformer in the substation. The second transformer transforms the medium voltage to a high voltage before the connection with the grid. To improve the quality of the electricity transmitted to the grid, there is an output filter installed between the inverter and the grid.
Another essential component in a PV plant is cables, which physically connect all components. Cable sizes are chosen to meet the required power transport. The cable length should be minimized for keeping energy losses and investment cost low [9].
System losses
There are several sources for losses in a PV plant. The inverter and transformer induce losses, as mentioned above. Losses for specific inverters can be found in the datasheet, otherwise received from the manufacturer. The PV modules are degrading with time, the magnitude of the degradation differs between modules, manufacturers and possibly the operation conditions. A more expensive PV module with high quality can be expected to have a lower degradation compared to a cheaper module.
New losses should be included when PV modules are connected and operated at a system level, one of these losses are due to a shared MPPT for modules with different operating conditions and therefore also different MPPs. This is called a mismatch loss, common causes are differences between the modules from the manufacturing, partial shading, differences in degradation and uneven soiling [11].
Numerous factors are affecting the soiling on the modules. There are aspects regarding the local conditions at the site, the design of the plant as well as factors coupled to environmental factors of naturally occurring phenomena. Factors varying with chosen location are dust, sand, particles, bird droppings, leaves, salt among many. Factors coupled with the environmental phenomenon is wind, snow, rain, etc.
Figure 2.7: A rooftop PV plant in Sweden during winter.
In a large research project conducted in the US with soiling losses measured from more than 200 large grid-connected PV plants in California, it was concluded that the daily system output was reduced by 0.2 % per day with a dry condition and no rain. Another conclusion was that rainfall of at least 20 mm rain is needed for cleaning the modules. This approach suggests a yearly soiling loss of 1.5 - 6.3 %, however, it is highly dependant on the site [13].
Typical system losses are presented in the report ”Analytical Monitoring of Grid-connected Photovoltaic Systems” published by IEA. Typical values stated in the report can be seen in Table 2.1:
Table 2.1: Typical losses in a PV plant [12]. Mismatch 0.98%
Diodes and connections 0.995% Wiring (DC + AC) 0.97% Soiling 0.95% System availability 0.98% Shading 0.95%
PV module degradation 0.5 % per year 2.1.2 Design of PV plant layout
There are several aspects to take into consideration when designing PV plant which is reliable and produces as much high-quality electricity as possible, to the lowest price possible. The main aspects that will be further presented are the panel placement relative to the sun, shading of the modules as well as the combination of modules and inverter capacities.
The first aspect is panel placement relative to the sun. The optimal tilt angle and orientation of the panels depend on location, local climate, and profile of the electricity pricing. A general rule for the elevation angle is that an elevation angle equal to the angel of the latitude maximizes the received insolation. However, the optimal elevation is site specific and should be analyzed with specific site data for solar irradiation [16]. Large PV plants, as well as rooftop systems, are normally built with the panels facing south in Europe [9]. The mounting system could have a tracking system with one or two degrees of freedom. These systems were more popular before the rapid cost reduction of PV modules. A two-axis tracking system may lead to a 25% increase of the electricity output, however, the cost of the tracking system and increased steel structure alter the potential monetary benefits. Although it might not be the last era of the tracking systems, as they may have a new role to fill in the future if there will be a large hourly difference in the electricity prices during a day. The tracking system can increase the production of a south-facing system during morning and evenings. This may be a cheaper solution than installing batteries. Another solution for increasing the hours of production is to design an east-west facing system.
The second aspect is panel placement relatively to objects. If the spacing between the rows is not large enough, near shading may occur. Another cause for near shading is surrounding obstacles, which can be the security fence for the plant, the housing for inverters, trees or any other objects at the site. Even though the near shading only affects a few modules, it may have a large impact on a system level due to a common MPP. Shading from obstacles should be eliminated to the largest extent possible. When it comes to shadowing between rows of modules, a rule of thumb is to design the spacing with the minimum distance that results in no module shading at noon of the winter solstice [9].
2.1.3 Performance
The system performance is a critical aspect of a PV plant project since it determines the power output of the plant. The predicted power output during the lifetime of the plant is crucial for determining the feasibility of the project since it is the main source of income. The design phase hosts the search for best performance in relation to low investment and lifetime costs.
for projects at new and unexplored sites. The data which is fed to the model is often interpolated from meteorological models of the ambient temperature and global horizontal irradiance. Since these are not the true conditions for the PV plant, determining the true values is stated to be one of the largest contributions of uncertainties in the modeled power output. Another large contributor of uncertainties is the fact that manufacturers stated module data is based on STC conditions, which seldom is the actual working conditions. STC conditions assume an irradiance of 1000 W/m2 and a cell temperature of 25◦C. When the data of the
ambient conditions are determined, the next challenge is to determine how it affects the performance of the modules [9].
One aspect to keep in mind is the variations of yearly irradiance, which can vary +/- 10% in Sweden. This variation directly translates to a change of the yearly power production. A PV plant placed in Sweden can be expected to have the performance of 800-1100 kWh/kWp. This includes the assumptions of panels due south and with an inclination angle between 30-50◦.
The module performance is affected by soiling on the panel, the extent of the soiling will be an assumption by the designer/person who does the modeling of the plant if the conditions of the site are not known in advance. The impact of soiling is most often stated as a percentage of reduction of yearly energy production. The recommended value to assume differ.
Depending on the type of soiling, the effect on the modules differs due to if it is an evenly distributed or partly shading from the soiling. Partly shading of a module can be the result of leaves or bird droppings. This can result in so-called hot spots on the module if there are no bypass diodes in the PV module [13]. Due to the long lifetime of the system, modeling of the power production must include the module degrada-tion. A low degradation of 0.2% per year results in a system degradation of 6% after 30 years. However, the degradation rate could be much higher depending on the chosen module. It is common to get a warranty for having 80% of the original power capacity of the module after 20-25 years. The lifetime of the modules can be expected to be at least 30 years [14].
2.1.4 Operation and maintenance
The operation and maintenance (O&M) strategy for PV plants has gone from being mainly corrective to being more advanced and preventive by focusing on avoiding component failure and thereby lost power production. This offers more long-term asset management and possibly a more optimized financial return. Having systematic maintenance, not only improves plant performance but also increases the safety of the system by reducing risks for accidents during maintenance as well as components igniting.
A more advanced O&M strategy is a condition-based way of structuring the maintenance. This includes preventative maintenance, detailed monitoring of the plant performance which can be used for prioritizing possible maintenance procedures. Corrective maintenance has to be prepared for unforeseen problems. Optimizing the O&M requires an analysis of the cost of the specific maintenance event and the resulting improvements of the system performance. The costs are highly dependent on the site and situation such as travel distance, labor cost, and the specific procedure. The cost of O&M is typically increasing with the age of the system. Companies that offer O&M as a service often increase the yearly fee with 2-4 percent due to the increased wear and tear of the system.
2.2
Grid connection
The grid connection is an important aspect of a PV plant project. This chapter will present the main features of the Swedish grid and how it is regulated, followed by information about how to connect a PV plant to the grid and the grid companies in Sweden.
2.2.1 The Swedish power grid
decision was made to transport power from the north to the metal industry located in the mid- and south part of Sweden. The next major step in the history of the Swedish power grid was to connect with the power grids in the neighboring countries [17, 18].
The Swedish grid is connected to the European power grid as well as the Nordic power grid. The Nordic power grid consists of the power grids in Sweden, Norway, Finland, and Denmark. The reference price for electricity in the Nordic power market is set by Nord Pool Spot. One service is setting the spot price for each hour in the upcoming 24-hour time period, based on all incoming bids for selling and buying electricity in the areas in the grid [27].
There are four different grid areas in Sweden, which constitutes four different bidding areas in the Nord pool spot market. The electricity price can differ between the four grid areas, due to restraints in the transmission lines between the areas. The higher power production in the north results in that the two upper grid areas, SE1 and SE2, has an overproduction. The high consumption in the southern part of Sweden results in that the two southern grid areas, SE3 and SE4, has a deficit of power production. Times with higher electricity prices in SE3-4 compared to SE1-2 is a sign of a limited capacity for transmitting electricity between areas [20]. An illustration of the four grid areas can be seen below, in Figure 2.8.
Figure 2.8: The four grid areas in Sweden.
which makes the reduces investment costs of lower voltage equipment more important than reduced losses. Different voltage levels are connected by transformers. The three levels are called the transmission network, the regional network, and the distribution network. The transmission network has the highest voltage levels, between 220-400 kV and serves the transport of the electricity from the large generations facilities to the demand. The regional network connects to the transmission network and has a voltage level between 30-130 kV. A regional network is restrained to a smaller area and has the purpose of connecting medium size production facilities and demands as well as transporting electricity to the distribution network. The distribution network is the smallest grids structure, which is connected to a regional network and has a voltage up to 20 kV. The regional- and distribution networks are owned and operated by numerous different grid companies. The transmission network is owned and operated by Svenska kraftn¨at, called Svk, which is a state-owned agency [27, 24].
The business of owning and operating the grid in Sweden is a monopoly, thereby it is highly regulated by laws, ordinances, and regulations which will be further presented in the next subsection. Owning, expanding and building a grid requires a so-called grid concession. Being certified as the only company with the legal right to having a grid in an area or a line comes with many responsibilities. A grid company must accept requests for connecting electric power production facilities during reasonable conditions [24].
2.2.2 Regulations
This chapter will present the main regulations regarding power production facilities and power grid networks. The main focus is directed to the Electricity law and grid codes.
The Electricity law in Sweden (Svenska: Ellagen) regulates the trade of electricity networks and electricity. The law was stated in 1997 and includes the production and usage of electricity as well as the transmission of electricity [21]. It has thirteen chapters, of which chapter two to four has specific importance for this report and will be further presented below. (For the complete content see Ellagen (1997)).
• Chapter 2 - Grid concession • Chapter 3 - Network activity • Chapter 4 - Tariffs
The following paragraphs will state some of the most important information from the three chapters. Grid concession
A high voltage transmission line cannot be built without a so-called grid concession. It also includes excava-tion, deforestation or other site preparing activities. There are two different grid concessions, one for a line and one for an area. Most commonly, the grid concession for an area allows transmission lines up to 20 kV. One main difference between the two, is that a grid concession for an area allows the grid company to build new lines within the limits of the maximum stated voltage level within the area, without receiving a new grid concession. However, the grid company must always consult other involved parties such as the municipality and landowners. For transmission lines with voltage levels higher than permitted in the permitted grid concession for the area, a grid concession for a line is needed. A grid concession for a line only includes that specific line and all expansions of the grid must have a new grid concession [21, 24].
The purpose of grid concessions are founded on the protection of common interests in our society, the right of any individual, ensuring safety for human and environment as well as implementing a scarce usage of resources [21].
Network activity
A company that owns a network is responsible for the operation, maintenance and if necessary, expansion of the network as well as connecting to other grids if applicable. The company also has the responsibility of ensuring the safety, reliability, and efficiency of its network.
A grid concession (both for line and area) includes the responsibility of connecting a power production facility to their grid within reasonable conditions and doing so within a fair time. If no special conditions apply, the connection should be done within 2 years from approval of the grid connection request. It also includes the responsibility of transferring electricity for other parties with good quality and measuring the transferred electricity. [Paragraph 8§ - 10§]
If a power production facility is located within a grid concession for the area and wants to connect to a grid with a concession for a line, the company with the grid concession for the area must give their approval. [Paragraph 8§]
A production facility has the right to receive compensations from the grid company, equal to the reduced electricity losses caused by the electricity from the production facility and the reduction of fees paid by the grid company to the regional/transmission network it is connected to [21]. [Paragraph 15§]
Tariffs
The earnings allowed for a grid company is regulated in the electricity law and is based on the projected costs and a reasonable rate of return. The tariffs must be objective and non-discriminating. This means that a grid company can neither subsidize nor have overprized tariffs for production from specific sources or technologies.
Grid tariffs are regulated differently regarding if it is a network in an area or line. A network with a grid concession for the area is not allowed to have differentiated tariffs depending on the location of the production facility. A regional network should have different tariffs for the different voltage levels to represent all their costs for the networks with the voltage level in question, tariffs may not be specifically developed regarding the distance between the connection point and the closest demand. [Paragraph 3§, 5§ and 8§]
A connection fee for connecting a production facility to the grid should be developed to cover the cost of each specific case. Specific rules apply to the connection of production facilities with a capacity equal to or smaller than 1.5 MW. The tariffs are only allowed to cover the grid companies costs for measuring, calculating and reporting values. [Paragraph 9§ - 10§]
Having a grid concession comes with the responsibility of giving a fast answer to questions about their tariffs as well as having them published [21]. [Paragraph 11§]
Grid codes
Since the grid is a complex structure, a document is published with common grid codes for the Nordic grid. Four codes are stated, the Operational Code, the Data exchange Code, the Planning Code, and the Connection Code. The Connection Code is of interest for this report and will be presented further. The idea with the Connection Code is to present basic rules for connecting a power production facility in a non-discriminating way that ensures continued secure operation of the Nordic grid. It is presented as the minimum technical requirements, the national requirements may be stricter. The technical requirements are stated for ensuring a high quality of the electricity. Specific regulations concerning how the power plant should operate during frequency and voltage variations in the grid, requirements for the generator and voltage characteristics [28].
2.2.3 Grid tariffs
There are several fees for being connected to the power grid. The fees can be divided into costs for the initial connection of the production facility and continuous transmission of electricity to the grid.
fee differs in Sweden. A customer that increases the losses pays for the electricity that cannot be sold and customers which reduces the losses get paid. [22, 24].
A connection fee is paid by a new customer to the grid company it is connecting to. the connection fee should cover the actions needed to be taken for enabling the requested connection. This may include installing new lines, transformers, and switches. In some cases, it may lead to reinforcing the grid at other places. The cost is highly dependant on the need for reinforcing the grid and the distance between point for connection and the chosen location for the production facility. Higher voltage levels often induce higher connection fees due to more expensive components [24].
The connection fee changes with each specific case. To get an approximate number to use, an example from connecting wind farms to the grid in Sweden is studied. A wind farm of a few MW is estimated to have a connection fee of 2.2 million SEK for connecting to the local grid. The distance to the grid is 5 km and a small reinforcement is needed. Another example of connecting a 15 MW wind farm to the regional grid results in a connection fee of 5.3 million SEK. This includes the component costs for an internal grid of 5 km which is owned by the wind farm company, three lines of 4 km each for connecting to the closest substation as well as some components needed for the substation.
2.2.4 Connecting a PV plant to the grid
Many aspects affect the connection of a PV plant to the power grid. Some of the main aspects are the capacity of the PV plant, the distance to the grid and the capacity of other connections of consumption and/or production close by the chosen point for connection. Even though the local conditions highly affects the connection of a power production facility to the grid, some general rules can be stated : a wind farm with a capacity smaller than 15 MW can be connected to the local grid, a capacity smaller than 300 MW can be connected to the regional grid and otherwise the connection should be made to the transmission grid. For a request of connection to the transmission grid to be approved, the production facility must have a capacity of at least 100 MW for connecting with 220 kV transmission line and at least 300 MW for connecting with a 400 kV network [23, 24].
The process of reaching a grid connection is described as typically following the steps : [24, 25] 1. Idea formed by a project company
2. Pre-feasibility study performed by the grid company, giving the project company an indicative price for the grid connection
3. Preparation of final tender, the final price and requirements are presented by the grid company 4. Start of project. The project company orders the new connection
5. Application for environmental permission as well as applications for new grid concession, if needed. Done by the grid company
6. Connection of the power production facility to the grid 7. Inspection
8. Operation
The document with the indicative price shall also include a sketch of the connection and if there would be any specific restrains for the project such as reinforcements, maximum capacities, and specific permits needed. The normal time for receiving the information is 40 working days for a grid with 20-30 kV, according to SOU 2008:13.
The next step in the grid connection process is that the project company orders a final tender for the connection from the grid company. The specific choices of equipment in the power plant are now important, if changed, the grid company must be informed and given the chance of updating the final tender. A stringent tender should not be ordered unless all the other permits are approved. This tender will be charged for and will then guarantee the grid connection until the stated expiring date. If the project is ordered, the cost of the stringent tender will be discounted in the connection fee. The time for receiving a stringent tender can be expected to be 60 working days for a grid with a voltage level of 20-30 kV according to SOU 2008:13. The stringent tender should include technical specifications, costs, the period of validity and if applicable, compensations for benefiting the grid [25].
A new grid concession may be needed for the line connecting the plant to the closest substation. The process of reaching the grid concession often requires revision before the energy agency has the required information to state their judgment. The process of getting a new grid concession has a mean time of one year [26]. The following timeline illustrates the steps and the approximate time for the process of getting a PV plant connected to the grid.
Figure 2.9: The process for connecting a new plant to the grid.
technical aspects covered in the agreement often includes the rated capacity and its variations, reactive power, slow voltage variations, flickering as well as resonance [24].
All connection points between a producer/consumer and the grid must have an assigned responsible person for balancing the power. Being responsible for balancing comes with the obligations of reporting forecasts of consumption and production at the connection point to Svk. It is possible to negotiate with the electricity supplier to take on the responsibility [27].
There are no specific requirements for connecting a PV plant to the grid, written in the Nordic grid codes published in 2007. Specific requirements are applied to hydropower, thermal power, and wind power. The document also includes general connection codes for connecting a production facility to a grid with a voltage of 110 kV or higher [28].
Grid owners in Sweden have the responsibility of sustaining an acceptable voltage in their grid, leading to the grid owners having specific requirements on the electricity from production facilities. Standard requirements and recommendations for connecting large production facilities to a regional grid and smaller production facilities to the local grid are published by Svensk energi. Neither one of the documents states any specific requirements for a PV plant. Some general points stated in the documents are that the production facilities should keep the reactive power to a minimum to reduce losses in the grid [25, 29].
As the large scale PV plants increase in number and installed capacity in Sweden, the impact from the PV plants increases on the power grid. Specific grid codes for PV power may be implemented in the Nordic grid, as it has been in Germany for the last decade. The German grid codes require PV plants to take an active role in supporting the grid during faults. Voltage control of the grid should be supported by the PV plant with feeding reactive power to the grid. A low voltage ride through (LVRT) ability is also required of the PV plant, different requirements apply for different conditions in the grid [30].
2.2.5 Physical constraints in the grid
Understanding the foundations of the grid, with its components, designed capacity, possibilities, and con-straints, is an important factor for choosing a spot in the grid which has the potential of connecting the PV plant without having to do enormous work with reinforcing the grid, which would induce large costs to the project. Two main aspects should be investigated, namely, if the physical components in the grid can receive the capacity in the new production facility as well as how the specific power production will affect the power flow in the grid.
The strength of a grid is one of the main aspects affecting the quality of the electricity in a grid. A strong grid has a reliable voltage level over time, meanwhile the voltage level in a weak grid is more easily affected. The strength of the grid is closely connected to the short circuit capacity of the grid.
By connecting a PV plant to a local grid, the direction of power flow may change due to the changed balance of consumption and production in the area. If this is the case, the components in the existing grid must be able to meet the new conditions.
Determining if an existing grid can receive the new production facility without reinforcement has to be evaluated by the grid company, by simulating different load and production scenarios with the prevailing grid parameters. It is difficult for PV plant developers to evaluate how their plant will affect the grid, which leads to some projects requiring an iterative process of finding a good site with a good spot in the grid to connect with.
2.2.6 Grid companies in Sweden
The transmission grid is owned by the government and operated by Svk, as stated before. Svk bears the responsibility of balancing the grid, which can be described as keeping a balance between production and imports with consumption and exports of electricity. Svk also has the responsibility for the safety of the operation of the Swedish grid as well as the responsibility for giving green certificates for production of renewable electricity [24, 27].
There are 58 regional grids in Sweden, the three main owners are ”Vattenfall eldistribution”, ”E.ON Eln¨at Sverige” and ”Ellevio AB” [31].
In a study of grid tariffs conducted during 2014, 93 % of the investigated grid companies had a yearly fee for being connected to the grid. The cost varied from no cost to 70 900 SEK, with a mean value of 15 000 SEK. It should be noted that the company which has the highest yearly fee of 70 900 SEK, has neither a power tariff nor an energy tariff. The majority of the large grid companies does not exceed the mean value for the cost. 77% of the companies have a cost related to the installed capacity, e.i. a power tariff with a mean value of 80 SEK/kW/year. 32 % of the companies have an energy tariff, with a rather low spread of the cost and a mean value of 0.0265 SEK/kWh. 23% of the companies have all three components included in their total tariff. However, most commonly the grid companies charge a tariff including the components of a fix yearly fee and a yearly fee for the rated power. The study concludes that there is an extreme difference when comparing the stated tariffs applied on a case of a 3 MW wind turbine connected to each grid, the difference between the highest and lowest cost was a factor of 600 [32].
2.3
Siting
Placing a PV plant in Sweden gives a project company a total of 40 731 117 hectares to chose from [64]. However, there is only a small fraction of the land that is possible to choose due to current usage, legislation, land cost, among many reasons. There are many aspects to take into consideration when choosing the site for a PV plant. Local conditions affect investment cost, the yield from the PV plant and social acceptance. The choice of the site highly affects the economy in the project.
Figure 2.11: Global solar irradiation map for Sweden.
The map could be used for narrowing down areas of interest for further search of land. When looking at a specific site, the irradiation can be investigated in detail by downloading data from SMHIs service called STR˚ANG, or some other service with site-specific irradiation data.
(a) Population density per square kilometer [34] (b) Installed capacity of PV plants in Sweden [35]
Figure 2.12: The two figures illustrates population density and installed PV capacity in Sweden. Land close to highly populated areas has a tenancy of being more expensive. The cost of renting land differs across the country, with approximately 10 times higher prices in the south compared to the northern part of Sweden [36]. The mean value for a yearly fee for renting agriculture land was 1726 SEK/ha during 2018. The prices have increased 30% the last decade, however, the prices have been rather stable since 2011 [37]. An important aspect to keep in mind is that the mean prices differ largely with the quality of the land as well as if the PV plant would compete with other possible usages of the land. To keep the cost of leasing land low, one should investigate marginal land and land with a dual purpose.
Figure 2.13: A site in Germany with both wind power and solar power operating in a close distance. Marginal land has many definitions, however, it is generally defined as land with low chances of having a profitable business due to its size, biophysical conditions or location as some examples. In a report presented by the Swedish Board of Agriculture (Jordbruksverket), it is concluded that half a million ha of agriculture and grazing land for animals are not used and that it is an overproduction of grazing. It is also concluded that most of the land is small separate areas [40, 41].
The report also includes an analysis of areas withdrawn from the database for subsidies of agriculture land as well as area which can not be used for agriculture. The largest reduction of subsidized agriculture land is located in the areas V¨astra G¨otaland and Sk˚ane, both located in the south of Sweden [40].
Another important aspect of the site selection is the existing power grid close by the site. Both the distance to the grid and finding a grid strong enough is important for keeping the cost for the grid connection low [41].
To simplify the logistics and reduce installation costs, the area should be flat as well as having an existing infrastructure close by.
2.4
Economics
This part presents the economic perspective of a PV plant project, which will be implemented later in the report. The economic model of calculating the LCOE of a project will be presented to start with, followed by a presentation of the parameters to include in the model for a PV plant located in Sweden. Expected prices to use in the models are presented as well.
2.4.1 Economic model
If a commercial grid-connected PV plant is to move on from the stage of a pre-feasibility study to implemen-tation, it is highly dependent on the economy in the project. There are initial investment costs, monthly fees as well as income for sold energy to take into consideration. The analysis naturally includes future aspects since it covers the lifetime of the PV plant. This requires assumptions of how the included variables changes with time. The analysis is also highly dependent on the actual power production which is simulated during the pre-feasibility stage.
The economic analysis can be presented as a Levelized Cost Of Energy, (LCOE), which represents the total cost per produced kWh during the lifetime of the project. The LCOE can be calculated as [42]:
where the life cycle cost is calculated as: i=N X i=1 Costi= CAP EX + i=N X i=1 OP EXi (1 + R)i −
V alue af ter lif etime
(1 + R)i (2)
and the power production during the life time of the plant is compensated with the system degradation as following: i=N X i=1 P roductioni= i=N X i=1
Initial production · (1 + degradation)i−1
(1 + R)i (3)
The parameter i represents the year, N the lifetime of the system and R represents the discount rate. The terms CAPEX and OPEX represent Capital Expenditures as well as Operational Expenditures, respectively. The CAPEX includes the initial costs for buying the plant and getting it up and running. The OPEX includes operation and maintenance costs as well as recurring costs of land lease, grid connection, etc. The production and system degradation is a technical question rather than economic. However, the accuracy of the simulation of performance highly affects the accuracy of the economic analysis. The model for simulating the performance of a plant will be presented later in the report.
There are different aspects to consider regarding an LCOE value, depending on the business model chosen for the specific PV plant. Regardless if the energy is sold on the spot market or sold by a PPA to a specific customer, there are some costs and incomes for the produced electricity.
If the electricity is sold on the spot market, the LCOE value should be equal or lower than the sum of the spot market price and the other sources of income and fees, which will be further explained in the next subchapter. If the chosen business model is PPA on the other hand, the spot market price is replaced by the price that the customer pays for electricity.
2.4.2 Expected prices
The total system cost for different types of PV systems has been bottom-up modeled by NREL (National Renewable Energy Laboratory) on a yearly basis for the last years. The total system price for a large scale PV plants (≥ 2 MW) with modules that are ground mounted with a fixed-tilt for the first quarter of 2018 was presented as 1.06 dollars per installed W. The price includes all system costs of components and installation as well as costs for the project development. The model is based on typical installation techniques and the prices for components are implemented from a developers point of view, meaning that it is the prices paid at the last stage instead of manufacturing costs [43].
One should notice that the CAPEX includes mounting and installation costs which differ between different countries. This implies that CAPEX should be adjusted or specific for each country. Data for the investment cost for specific projects can often be difficult to find, however, it is available for the two largest PV plants in Sweden. The total investment cost for the PV plant in S¨ave was 45 million SEK, leading to a CAPEX of 8200 SEK/kWp. The PV plant in Varberg had a total investment cost of 24 million SEK, resulting in a CAPEX of 8900 SEK/kWp. Comparing with the CAPEX presented above, from NREL, using a currency exchange rate of 8.2 SEK per dollar as an approximate representation for the conversion rate during 2018 when the NREL report was published. The resulting CAPEX in SEK from the NREL report is 8700 SEK/kWp which is in between the values presented for the two largest PV plants in Sweden [44].
The total cost can be divided further into separate components. This is valuable information for the internal cost optimization of the PV system during the design phase. The following section will further describe costs for central inverter and PV modules.
A central inverter for a large scale PV plant with fixed tilt was presented as having a mean value cost of 0.06 USD/Wdc with an inverter load ratio of 1.36, during 2018. The minimum and maximum values were presented as approximately 0.02 USD/Wac and 0.12 USD/Wac respectively [43].
chain costs for shipping and handling as well as module related sales taxes. Using the ”soft costs” presented for 2018 in combination with the global spot price, the total price adds up to 0.41 USD/Wdc [43].
The operation and maintenance, including both preventative and corrective maintenance, are modeled by NREL with the perspective of costs. Corrective maintenance includes both labor and component costs. Com-ponent failure is derived from a Weibull-distribution for calculating the probability of failure per comCom-ponent and year. The O&M cost was stated as 9.1 USD/kW/year during 2018 [43].
Regardless of the business case for the PV plant, there are several costs/incomes for the produced electricity, some of which are changing in time and some in space. They will be stated and then further explained below:
• Fee/compensation for energy fed to the grid • Corporation tax
• Green certificates
Fees and/or compensation from the grid company for feeding energy to the grid differs across the country, between the different grid companies as well as it changes over time. This was explained in section 2.2.6, however, for specific values of the fees and compensations one must look at the specific grid company. There is a corporate tax in Sweden, the same as in many other countries. The corporate tax applies to the profits made by a company. The tax is decreasing due to governmental decisions, from 22% from 2013 to 2018 to 21.4% during 2019 and it will decrease to 20.6% during 2021 [45].
Green certificates are received for producing renewable energy in Sweden. This system was introduced in 2012 in Sweden and Norway as a support system for increasing renewable power production. The producers receive green certificates and electricity retailers (among others) have an obliged share of green certificates per sold unit of energy to sustain. The price of a green certificate is decided by the market and is highly fluctuating on a monthly basis. The mean price was 0.03 SEK/kWh during 2017 [27, 46, 47, 48].
There are suggestions for ending green certificates by the year 2030. With the great amount of renewable power recently installed, there will be a surplus of production of green certificates, which will lead to a reduction of its value [49].
A system that is easily mixed up with the green certificates, is the system for ensuring the origin of electricity. Certificates are given for each MW of produced electricity, stating its origin. The system ensures that sold electricity has the origin that the merchant states. There are two main ways for these certificates to be used, an electricity supplier selling renewable electricity from solar energy (as one example) must either produce the same amount of solar electricity as it is selling. In this case, the electricity supplier will receive certificates for the origin of the solar electricity which they have to annul when it is sold. The second case is that the electricity supplier sells solar electricity and instead of producing it themselves, they buy the certificates from a producer of solar electricity. This case also ends up with the electricity supplier having annulled the certificates after selling the electricity. The certificates are sold on a spot market to an average price of 0.01 SEK/kWh.
Commercial plants feeding electricity to the grid without a specific customer sells the electricity to a price decided by the spot market at Nord pool, which is changing with time and to a small extent between the four grid areas in Sweden. A price that may be used as a representative for the near future (of a few years) is the yearly mean value from 2017, which was 0.3 SEK/kWh [50].
