Multi-objective optimization tools using genetic algorithms (GAs) are being increasingly used for improving building performances and sustainability. However, few research studies focus on district-scale solutions. In the present project, a multi-objective optimization method using genetic algorithms was applied in order to help decision makers find the optimal energy mix of a district energy system in the preliminary design phase.   A case study consisting of the new campus Albano in Stockholm (comprising lecture buildings and student residences) was used for the analysis. A wide range of energy systems was included as a design variable: wind turbines, solar thermal collectors and photovoltaic cells, ground-source heat pumps, biomass boilers, combined cooling, heating and power, district heating and district cooling. The energy provided by the chosen technologies and the district energy balances are simulated on an annual basis using a steady-state method with an hourly resolution.   Three objectives functions were to be minimized: (1) the life-cycle costs; (2) the greenhouse gas emissions; and (3) the annual non-renewable primary energy consumption of the district. The optimization process was implemented on MOBO, a multi-objective optimization tool based on genetic algorithms.   The findings include understanding the trade-offs among the three objectives and a selection of alternatives of energy supply systems to be further investigated in the detailed design phase.

<p>Climate change is recognized as one of the most significant challenge of the 21st century. Anthropogenic activities have led to a substantial increase in greenhouse gases (GHGs) since the Industrial Revolution, with the energy sector being one the biggest contributors globally. The energy sector is now facing unique challenges not only due to decarbonization goals but also due to increased risks of climate extremes under climate change. </p><p>This dissertation focuses on leveraging machine learning, specifically utilizing unstructured data such as images, to address many of the unprecedented challenges faced by the energy systems. The dissertation begins (Chapter 1) by providing an overview of the risks posed by climate change to modern energy systems. It then explains how machine learning applications can help with addressing these risks. By harnessing the power of machine learning and unstructured data, this research aims to contribute to the development of more resilient and sustainable energy systems, as described briefly below. </p><p>Accurate forecasting of generation is essential for mitigating the risks associated with the increased penetration of intermittent and non-dispatchable variable renewable energy (VRE). In Chapters 2 and 3, deep learning techniques are proposed to predict solar irradiance, a crucial factor in solar energy generation, in order to address the uncertainty inherent in solar energy. Specifically, Chapter 2 introduces a cost-efficient fully exogenous solar irradiance forecasting model that effectively incorporates atmospheric cloud dynamics using satellite imagery. Building upon the work of Chapter 2, Chapter 3 extends the model to a fully probabilistic framework that not only forecasts the future point value of irradiance but also quantifies the uncertainty of the prediction. This is particularly important in the context of energy systems, as it relates to high-risk decision making.</p><p>While the energy system is a major contributor to GHG emissions, it is also vulnerable to climate change risks. Given the essential role of energy systems infrastructure in modern society, ensuring reliable and sustainable operations is of utmost importance. However, our understanding of reliability analysis in electricity transmission networks is limited due to the lack of access to large-scale transmission network topology datasets. Previous research has mostly relied on proxy or synthetic datasets. Chapter 4 addresses this research gap by proposing a novel deep learning-based object detection method that utilizes satellite images to construct a comprehensive large-scale transmission network dataset.</p>

Den svenska regeringen har som mål att Sverige år 2045 inte ska ha några nettoutsläpp av växthusgaser, och därefter sträva efter negativa utsläpp för att motverka den rådande klimatförändringen. Detta då människans användande av fossila bränslen anses vara den största orsaken till de globala klimatförändringarna. För att i framtiden kunna minska människans påverkan är det viktigt att förnybara energikällor fasar ut de fossila, eftersom en markant minskad energikonsumtion inte är rimlig ur ett samhällsekonomiskt perspektiv. En växande befolkningsmängd och därmed ett ökat energibehov ligger framför oss samtidigt som Sverige har som mål att vara ett föregångsland när det handlar om utfasningen av fossila bränslen. I landet finns det goda geografiska förutsättningar som kan bidra till energiproduktionen på flera olika sätt. Sveriges stora areal i förhållande till befolkningsmängden är också en av fördelarna varför skogsbruk anses vara en av framtidens viktiga näringar och det är därmed av intresse att undersöka dess potential.  I denna studie har potentialen för skogen som energikälla genom brukande av outnyttjad åkermark studerats ingående på en nationell nivå för att kunna undersöka hur stor del av framtidens energisystem i Sverige som kan baseras på inhemsk biomassa. Dessutom har även energigrödan salix behandlats för att visa på ett alternativ till andra trädtyper. Resultaten visar att det finns många komplexa faktorer som måste tas i hänsyn för att odlingen av biomassa för energiproduktion i Sverige ska vara hållbar, även med en ökad intensitet och produktion. Det kommer att krävas noggrann planering och bakomliggande forskning som bidrar till att andra faktorer i naturen inte tar skada av skogsbruket, åtminstone så att denna påverkan minskas i största möjliga mån. Detta genom styrmedel i form av incitament för forskning inom teknisk utveckling samt regler för att kunna styra producenters och konsumenters attityder och agerande gällande hushållning av energi och naturresurser.  Baserat på tre scenarier där energimängd beräknades utifrån tillgänglig åkermarksarea bedöms inhemska biobränslen inte kunna ersätta dagens energianvändning av fossila bränslen. Scenariot med störst bidrag till energisystemet skulle endast ersätta cirka 7 procent av dagens fossila bränsleanvändning i Sverige. Biobränslen kan däremot bidra till mer hållbara alternativ inom bland annat transportsektorn och tillverkningsindustrin. De förnybara biobränslena bidrar framförallt till minskade utsläpp av växthusgaser, samtidigt som en odling av biomassa och framställningen av dessa förnybara bränslen kan innebära negativ påverkan på den lokala miljön. En omställning från ett fossilbaserat system till ett system grundat på förnybara energikällor medför att miljöbelastningen kan komma att överföras från att drabba vissa miljömål till att påverka andra. Det innebär alltså att det behövs en balans och prioritering mellan olika miljömål. Klimatmålet att Sverige ska vara fossilfritt år 2045 är däremot inte omöjligt om fler förnybara energikällor inkluderas samt en viss mängd importeras. / The Swedish Government has set a goal that Sweden will have zero net emissions of greenhouse gases by the year of 2045, and then strive for negative emissions to counter the current climate changes. This is because human usage of fossil fuels is considered to be the main cause of the global climate change. To reduce the human impact in the future it is essential that renewable sources of energy will phase out the fossil ones, since a significantly decrease of energy consumption is not likely from a socioeconomic perspective. A growing population and therefore an increased need of energy lies ahead of us, while Sweden aims to be a leading country regarding phasing out fossil fuels. The country has suitable geographical conditions that in several ways can contribute to the production of energy. Sweden’s large area in relation to population is also one of the reasons why forestry is considered to be one of the most important industries in the future and it is of interest to investigate its potential.  In this study, the potential for the forest as an energy source through the use of unused arable land has been studied extensively at a national level in order to investigate how much of the future Swedish energy system can be based on domestic biomass. In addition, the energy crop salix has been included to show an alternative to other tree types. The results show that there are multiple complex factors that need to be taken into account for the cultivation of biomass to be considered sustainable, even with increased intensity and production. There will a need for careful planning and underlying research that will prevent the forestry from harming other factors in the nature, at least to reduce its impact as much as possible. This can be done with political instruments in terms of incentives for research in technological development and rules for guiding the attitudes and actions of producers and consumers regarding energy and natural resource management.  Based on three scenarios where the amount of energy was calculated from the available arable land area, domestic biofuels are not expected to be able to replace today’s energy use of fossil fuels. The scenario with the greatest contribution to the energy system would only replace about 7 percent of today’s fossil fuel usage in Sweden. Biofuels can, on the other hand, contribute to more sustainable alternatives in the transport and manufacturing sectors. Renewable biofuels contribute primarily to reduced greenhouse gas emissions, while the cultivation of biomass and the production of these renewable fuels can adversely affect the local environment. A transition from a fossil-based system to a system based on renewable energy means that the environmental impact can be transferred from affecting certain environmental goals to affect others. Therefore, there is a need for a balance and prioritization between different environmental goals. The climate target that Sweden will be fossil-free in 2045 might be possible if more renewable sources of energy are included and a certain amount is imported.

Yes / The cost of electricity produced by thermal power plants in Republic of Djibouti is relatively high at about $0.32/ kWh. This is due to its dependence on imported oil coupled with fluctuating oil prices. Consequently, the customer pays a high electricity bill. However, Djibouti is endowed with indigenous renewable energy resources such as a good solar irradiance of 5.92 kWh/ m2 day, a potential geothermal energy estimated up to 1000 MW, and few sites with annual wind speed higher than 6 m/s. The goal of this paper is, therefore, to assess an economic evaluation of different grid connected hybrid renewable energy systems to a residential urban house located in Tadjourah city (11.7913◦ N, 42.8796◦ E) in the North-Eastern part of Djibouti to reduce the cost of electricity from the grid. To reach this objective, a powerful software tool called HOMER (Hybrid Optimization Model for Electric Renewables) has been used to find the optimum hybrid energy system using real wind and solar irradiation data. The results obtained from this study show that the best economical suited combination of hybrid renewable energy system is a PV-Wind grid connected system. This study shows also that potentially the indigenous renewable energy contribution, in Tadjourah, can be as much as 77 % with 47 % of solar and 30% of Wind energy. The Net Present Cost, the Levelized Cost of Energy, and the operating cost of the optimal HRES are $337, $0.002/kWh and $1,025/year, respectively. When compared with the average cost of grid-only connection of $0.32/kWh, the optimal hybrid renewable energy system is more economical and will save 51 % of the cost that the customer must pay when using only the electricity from the grid.

This master thesis investigates the impact of battery energy storage utilization in the ancillary services market on a local electricity grid. Specifically, the study examines the technical impact of battery energy storage installations and how grid operators should handle these types of installations in the future. To achieve this goal, battery energy storage systems simulations using MatLab software based on historical frequency data have been conducted creating an operation scheme for the battery. The battery simulation was applied to the electrical systems consumption and production data, conducting a whole year power flow simulation investigating the effects on peak power demand both for the customer and local grid owner. The worst-case scenarios were also investigated to assure that the grid could withstand the change in production and consumption at any time. The results indicate that such an installation would increase the peak power demand for the local grid owner by a maximum of 12 kW for one week but in general, it would result in a smaller increment and some small decrement. For the customer, the maximum impact is greater both in absolute numbers and in proportion to the peak power demand before with an increment of 16 kW, but in general, there will not be any impact on the effect tariff at all. The results of the worst-case scenario simulations showed that during backup power supply at high load timestamps in combination with FCR-D down there would be contingencies in one cable connection between two nodes, causing the cable current to increase above the rated current. To allow such an installment the local grid owner would need to limit the battery capacity and increase their billings to the customer, expand the dimension of the system or even, increase their effect tariffs to always ensure a stable electricity supply and stable financial balance.

Zusammenfassung Um die ehrgeizigen politischen Ziele zur Reduzierung der Treibhausgasemissionen im Stromsektor zu erreichen, stimmen alle relevanten Energieszenarien überein, dass Deutschland kurz- bis mittelfristig bis 2035 seine Kapazitäten zur Erzeugung erneuerbarer Energien massiv ausbauen muss. Deutschland ist dabei wie viele andere Länder auch stark von fluktuierenden erneuerbaren Energiequellen (fEE) abhängig, insbesondere von der Wind- und Solarenergie. Die Spezifika der Stromerzeugung von fEE stellen neue und besondere Herausforderungen an ein zuverlässiges Stromversorgungssystem der Zukunft. Entsprechend hat die Erforschung der technischen Optionen bei der Integration großer Anteile von fEE in das Stromnetz in den letzten Jahren stark an Interesse gewonnen. Allerdings scheinen Energieszenarien die mit der schnellen technologischen Entwicklung einhergehenden Integrationsoptionen bisher nicht korrekt abzubilden. In der vorliegenden kumulativen Dissertation wurden ausgewählte technische Optionen für die Integration erneuerbarer Energiequellen in das Stromnetz im Rahmen einer Fallstudie für Deutschland sowie ausgewählter Übertragungsnetze in Deutschland untersucht. Zur Identifizierung und Bewertung der Integrationsmöglichkeiten, widmete sich die Arbeit den vielversprechendsten technischen Integrationsoptionen in Form von i.) systemfreundliche Auslegung von Wind- und Solaranalgen; ii.) optimale Kapazitätsanteile von Wind- und Solaranlagen, iii.) der räumlichen Allokation und Bewertung von Windenergieanlagen in herkömmlicher als auch systemfreundlicher Auslegung; iv.) und dem Beitrag welchen die flexible Stromerzeugung aus Bioenergie als Ergänzung zu steigenden Anteilen an fEE erbringen kann. Es wurde ein Methodenmix zur Beantwortung dieser Forschungsfragen genutzt, der von der numerischen Optimierung auf Basis von Zeitreihendaten über die räumliche Potenzialkartierung und Allokation bis hin zur multikriteriellen Entscheidungsanalyse reicht. Die Ergebnisse zeigen wie der Übergang zu einem von hohen Anteilen an vRES gekennzeichneten Stromversorgungssystem erleichtert werden kann. Darunter Möglichkeiten zur Beschleunigung des Umstiegs auf erneuerbare Energien mit deutlich reduzierten Erzeugungskapazitäten von Wind- und Solaranlagen, weniger negative Residuallasten und negativer residualer Energie, verbesserte Sektorenkopplung und die Potenziale der flexiblen Stromerzeugung aus Bioenergie als Ergänzung zu fEE.:Table of Contents Abstract Zusammenfassung Acknowledgements List of Publications List of Acronyms Table of Contents I. Introductory chapters 1. Introduction 1.1. Background 1.2. vRES in energy scenarios 1.3. Technical developments and options for the integration of vRES 2. Research questions 3. Methods applied in this PhD thesis 4. Discussion and conclusion 4.1. Summary of the main findings 4.2. Transferability of results and methods 4.3. Relevance and outreach 5. Appendix 6. Literature 7. Appended publications and the individual contribution to the publications 8. Curriculum Vitae (deleted) 9. Selbstständigkeitserklärung / Abstract: For Germany to achieve its ambitious political targets for the reduction of greenhouse gas emissions in the electricity sector, major energy scenarios and reports project that the country will have to expand its renewable power generation capacities massively by 2035. As is the case for many countries, Germany will have to heavily rely on variable renewable energy sources (vRES), especially wind and solar photovoltaics. The characteristics of power production from vRES pose challenges for a stable and reliable future power supply system. Accordingly, the research into the technical challenges of integrating large shares of vRES into the power system has therefore attracted much interest in recent years; however, major energy scenarios seem to not cover integration options associated with the fast development of vRES correctly and lag behind the fast development in renewable energy technology. In this cumulative thesis, selected technical options for the integration of renewable energy sources into the power supply system have been investigated in a case study of Germany and a selected transmission system in Germany. To identify and assess these emerging integration options, the research in this PhD thesis covers the most promising technical options for the integration of vRES in the form of i) system-friendly layouts of wind and solar PV; ii) optimal capacity mixes of vRES; iii) the spatial allocation of wind turbines and the impact assessment of wind turbine allocation; and iv) the contribution of flexible power generation from biomass to complement vRES. Therefore, a mix of methods has been applied, ranging from numerical optimization based on time series data, GIS potential mapping and allocation including a multi-criterial decision analysis. The results show how the investigated options can facilitate the transition for a power supply system dominated by high shares of vRES in the near to medium term. A faster energy transition with significantly reduced overall vRES power generation capacities, less Excess Energy (EE) generation, improved cross-sectorial energy provision and flexible bioenergy as a complement to vRES are the major findings of the investigated options in this thesis.:Table of Contents Abstract Zusammenfassung Acknowledgements List of Publications List of Acronyms Table of Contents I. Introductory chapters 1. Introduction 1.1. Background 1.2. vRES in energy scenarios 1.3. Technical developments and options for the integration of vRES 2. Research questions 3. Methods applied in this PhD thesis 4. Discussion and conclusion 4.1. Summary of the main findings 4.2. Transferability of results and methods 4.3. Relevance and outreach 5. Appendix 6. Literature 7. Appended publications and the individual contribution to the publications 8. Curriculum Vitae (deleted) 9. Selbstständigkeitserklärung

The 2050 national energy scenarios as planned by the DECC, academia and industry specify a range of different decarbonised supply side technologies combined with the electrification of transportation and heating. Little attention is paid to the household demand side; indeed within many scenarios a high degree of domestic Demand Side Management (DSM) is implicit if the National Grid is to maintain supply-demand balance. A top-down, bottom-up hybrid model named Shed-able Household Energy Demand (SHED) has been developed and the results of which presented within this thesis. SHED models six published national energy scenarios, including three from the Department for Energy and Climate Change, in order to provide a broad coverage of the possible energy scenario landscape. The objective of which is to quantify the required changes in current household energy demand patterns via DSM, as are implicit under these highly electricity dominated scenarios, in order to maintain electrical supply-demand balance at the national level. The frequency and magnitude of these required household DSM responses is quantified. SHED performs this by modelling eleven years of supply-demand dynamics on the hourly time step, based on the assumptions of the published energy scenarios as well as weather data from around 150 weather stations around the UK and National Grid historic electricity demand data. The bottom-up component of SHED is populated by 1,000 households hourly gas and electricity demand data from a recently released dataset from a smart metering trial in Ireland. This aggregate pool of households enables national domestic DSM dynamics to be disaggregated to the aggregate household level. Using household classifications developed by the Office for National Statistics three typical ' households are identified within the aggregate pool and algorithms developed to investigate the possible required responses from these three households. SHED is the first model of its kind to connect national energy scenarios to the implications these scenarios may have on households consumption of energy at a high temporal resolution. The analysis of the top-down scenario modelling shows significant periods where electrical demand exceeds supply within all scenarios, within many scenarios instances exist where the deficit is unserviceable due to lack of sufficient spare capacity either side of the deficit period. Considering the level of participation required within the modelled scenarios in order to balance the electricity system and the current lack in understanding of smart metering and Time-Of-Use (TOU) tariffs within households, it would seem there is a disconnect between the electricity system being planned, the role this system expects of households and the role households are willing to play.

Approved for public release; distribution is unlimited / This thesis explores the causes of Russian foreign policy towards Georgia. It argues that the Russian Federation continues to pursue a policy which weakens the sovereignty of the Caucasus. The main priority of this thesis is to identify why the Russian Federation seems to be pursuing a set of policies that economically and politically weaken the sovereignty of Georgia. Therefore, this thesis examines the forces and factors of Russian domestic politics that drive Russian national interests towards the Caucasus. The analysis focuses on one particular issue-area: the role of the economic elite in shaping Russia's domestic and foreign policies vis-a-vis the state in the electricity sector. In focusing on the energy policies of the Russian Federation, this thesis reveals the negative consequences for Georgia's sovereignty that result from a strong Russian influence in the region. This thesis analyzes how Russian national interests towards Georgia challenge the latter to establish autonomous decisionmaking with regard to its foreign policy and to exercise its own authority through an exclusive competence in internal affairs of the state. In conclusion, this thesis offers policy prescriptions on how Georgia might best preserve its sovereignty with respect to the Russian Federation in terms of energy dependency. / Civilian, Ministry of Defense, Georgia

Skolfastigheter is a municipality owned real estate company that manages most of the buildings used for lower education in Uppsala. The company is working in line with the environmental goals of the municipality by installing photovoltaic systems in schools and other educational buildings. Skolfastigheter are planning to install a photovoltaic system in a school in Stenhagen. The purpose of this study is to optimally design the proposed system. The system will be maximized, which in this study entails that the modules will be placed on every part of the roof where the insolation is sufficient. The system will also be grid connected. The design process includes finding an optimal placement of the modules, matching them with a suitable inverter bank and evaluating the potential of a battery storage. Economic aspects such as taxes, subsidies and electricity prices are taken into account when the system is simulated and analyzed. A sensitivity analysis is carried out to evaluate how the capacity of a battery bank affects the self-consumption, self-sufficiency and cost of the system. It is concluded that the optimal system has a total peak power of almost 600 kW and a net present value of 826 TSEK, meaning that it would be a profitable investment. A battery bank is excluded from the optimal design, since increasing the capacity of the bank steadily decreased the net present value and only marginally increased the self-consumption and self-sufficiency of the system.

District heating is today the most common way of providing a building with heat and hot water in Sweden. It is an environmentally friendly product mostly used with renewable fuel. However, at power peaks most companies use production units that are more expensive and worse for the environment and should therefore be avoided as much as possible. This can be done with a method called load management. When a power peak occurs, the heat supply to buildings connected to the district heating system can be temporarily reduced. The heat energy can later be returned when the heat demand is lower. Thanks to the heat inertia of the buildings, the indoor temperature will not fall within the time frame for the load management. Historical data has been analysed to identify when and why power peaks occur in the district heating network. Power peaks throughout the district heating network have proved difficult to identify. However, for individual consumers clear patterns of power peaks have emerged. These power peaks mainly occur because of large use of hot water but also because of the shifting outdoor temperature. In order to see how the production cost would differ from the actual outcome load management was applied for Gävle's district heating 2018. The load management was calculated manually by identifying the most expensive production unit on an hourly basis. If a cheaper production unit had the potential to deliver higher power the next hour, the production was moved to the cheaper production unit. The process was repeated for each hour during 2018. After carrying out load management for Gävle's district heating network, 1 457 MWh had been shifted to a cheaper production unit. This resulted in a financial saving of 1,0 % of the total production cost. The environmental savings showed a reduction from 6.1 to 5.9 g CO2eq /kWh a total of 197 tonne CO2eq. In the exact same way, a load management was performed for a scenario where Gävle and Sandviken's district heating network were connected. The gain for a load management with Sandviken will be considerably larger, a reduced production cost of 3.6 % is possible. The environmental savings showed a reduction from 8.4 to 7.8 CO2eq /kWh a total of 575 tonne CO2eq. For future efficient load management, buildings should be divided into different classes depending on the building's time constant. User patterns for the entire district heating network have proved difficult to detect. Artificial intelligence can be an option for short-term forecasting of the power output / Fjärrvärme är idag det vanligaste sättet att förse en bostad med värme och tappvarmvatten i Sverige. Fjärrvärmen är ofta en miljövänlig produkt som kan produceras av till exempel biobränsle- och avfallseldade kraftvärmeverk eller spillvärme från industrier. Vid tillfälligt högt effektbehov, effekttoppar, använder sig merparten av bolagen av dyrare produktionsenheter med större miljöpåverkan. Dyrare produktionsenheter bör undvikas i största möjliga mån och i detta syfte används metoden laststyrning. Vid en effekttopp kan värmetillförseln till byggnader sänkas temporärt för att återföras några timmar senare när effektbehovet är lägre. Tack vare värmetrögheten i byggnaderna bör inomhustemperaturen inte sjunka inom tidsramen för laststyrning. Statistik från Gävles fjärrvärmanvändning på timbasis under 2018 har analyserats för att identifiera när och varför effekttoppar sker. Effekttoppar i hela fjärrvärmenätet har visat sig svåra att identifiera. På lokal nivå har däremot tydliga mönster för effekttoppar framkommit. Dessa effekttoppar beror till största del av tappvarmvattenanvändning men även förändringar i utomhustemperaturen. För att se hur produktion och last kunde skiljt sig från det verkliga utfallet tillämpades laststyrning för Gävles fjärrvärmeproduktion 2018. Laststyrningen beräknades manuellt genom att den dyraste produktionsenheten identifierades på timbasis. Om en billigare produktionsenhet hade potential att leverera högre effekt nästkommande timmar försköts produktionen. Därefter upprepades processen för varje timme under 2018. Efter utförd laststyrning för Gävles fjärrvärmenät hade ca 1 457 MWh förskjutits till en billigare produktionsenhet. Det gav en ekonomisk besparing på 1,0 % av Gävles totala produktionskostnad. Den miljömässiga besparingen visade på en sänkning från 6,1 till 5,9 [g CO2ekv /kWh] sammanlagt 197 ton CO2ekv. På samma sätt utfördes en laststyrning för ett scenario där Gävle och Sandvikens fjärrvärmenät sammankopplats. Vinsten för en laststyrning med Sandviken blev betydligt större med en minskad produktionskostnaden på 3,6 %. Den miljömässiga påverkan sjönk från 8,4 till 7,8 g CO2ekv /kWh sammanlagt 575 ton CO2ekv. För en framtida effektiv laststyrning bör byggnader delas in i olika klasser beroende på byggnadens tidskonstant. Användarmönster för hela fjärrvärmenätet har visat sig svårt att identifiera. Artificiell intelligens kan vara ett alternativ i framtiden för att prognostisera effektuttaget