Reconfigurable Intelligent Surfaces (RIS) have emerged as a promising technology to enhance the performance of next-generation wireless networks, particularly in terms of increasing coverage and enhancing user throughput. However, effective radio resource management for RIS-aided mobile networks introduces significant complexity due to the high dimensionality and the interdependency of these tasks.

This thesis proposes a solution that leverages the machine learning-based framework of Graph Neural Networks (GNNs) to jointly optimize radio resource management tasks in a multi-user scenario.
The proposed model is capable of performing user scheduling, beamforming, and RIS configuration based on full channel information that maximizes proportional fairness. Additionally, the model also supports implicit power allocation.

A comprehensive simulation environment emulating a dense urban mobile network is developed to train and test the model. The performance across various mobile network deployments is evaluated and compared to an accurate existing method.
Results demonstrate that the proposed GNN-based solution manages to achieve a large fraction of the user throughput gain achieved using the existing method in significantly less computation time, showcasing its potential for real-time radio resource management in future 6G networks.

Distribution Network Operators (DNOs) are confronted with a significant challenge to expand the capacity of the electricity distribution grid to facilitate the energy transition. Grid expansion planning for the distribution grid is a complex problem with many constraints and objectives. Earlier research has focused on metaheuristics such as genetic algorithms to optimise grid expansion designs in terms of resolved congestions, redundancy and costs. However, this algorithm is not used in practice by its intended users; grid architects. This research first identifies why the algorithm is not yet used and proposes several ideas to improve user adoption. Then a technical implementation of sketch-based optimisation using a novel shape similarity measure is presented. We investigate its behaviour with a realistic case study. Furthermore, a qualitative user study is done to evaluate the potential impact of sketch-based optimisation on distribution grid expansion planning. We conclude that the novel shape similarity measure enables sketch-based optimisation of distribution grid expansion planning and that it has potential to accelerate electricity grid design processes. Alliander, the largest DNO of the Netherlands, will implement sketch-based optimisation in their distribution grid design application later this year, making it available to 20+ grid architects to accelerate their design process.

The management of electric vehicle (EV) batteries involves accurately estimating their state-of-charge (SOC), which indicates remaining usable energy. Precise SOC estimation extends battery life, increases usable capacity, and enhances vehicle performance by allowing a greater depth of discharge without increasing battery weight or size. Traditional SOC estimation via Coulomb counting accumulates errors and requires correction using the Open-Circuit-Voltage (OCV). Modern silicon graphite anodes exhibit voltage hysteresis, making accurate SOC determination difficult. Introducing a hysteresis factor, ranging from -1 to 1, helps interpret OCV values within this hysteresis region.

This thesis proposes using a black box approach and machine learning models to predict the hysteresis factor for SOC correction more accurately. The research aims to enhance SOC accuracy in real-world EV environments, beyond laboratory settings. Improved SOC estimation through machine learning could significantly advance battery management systems (BMS), yielding both economic and environmental benefits by increasing EV efficiency. The study addresses the existing gaps in SOC estimation for voltage hysteresis, considering the vehicle as a system by incorporating driving cycles, BMS technical limitations, and cell chemistry. By improving EV reliability and efficiency, this research aims to promote broader acceptance and contribute to a sustainable future in personal transportation.

This research is carried out in collaboration with Vattenfall, to investigate the performance of a 76-turbine onshore wind farm. Since the start of operation, the wind farm has not been reaching P50 production estimates. Over the first five years of operation, there has been a big average gap between Annual Energy Production (AEP) and long-term AEP P50 estimates. This thesis gives insights into the physical causes of this apparent underperformance. Conventional 10 minutes averaged data as well as data with a higher resolution, named high-frequency data, has been available for this purpose. Therefore, the research question to be answered is: “Can the underlying physical causes for the underperformance of the Pen y Cymoedd wind farm be found by the use of 10 minutes averaged and/or high-frequency data?”.

Answers to this research question were found by following a set-up methodology, partially based on International Electrotechnical Commission (IEC) standards and with an added purpose for high-frequency data. This analysis was carried out for 14 turbines divided over 3 turbine clusters, with each turbine cluster accompanied by a MET mast, performing independent site measurements. The analysis only considers unwaked turbine sectors. After filtering and correlating the data sets, a cloud analysis and overall & directional performance analysis were carried out to understand the influence of the site as well as the influence of turbine controls to underperformance. The cloud analysis investigated deviating turbine behaviour, where the overall & directional performance analysis focused on the influence of the site on turbine performance.

The set-up methodology and analysis led to the following findings: on average, after data filtering, wind turbines perform 95% compared to the performance numbers the manufacturer delivers. There is a big difference in performance between the two turbine types at the site. Overall, the turbulence intensity (TI) class A turbines are performing better than the TI class B turbines. From the overall & directional performance analysis, bad performance (>-5% compared to the warranted performance) was seen for sectors with a complex orography, as well as for sectors with few elevation deviation or forestry. The cloud analysis revealed performance improvement potential for multiple turbine control regions. The high-frequency data gave additional unique insights in turbine performance, only visible in higher resolution. Moreover, it clarified turbine behaviour visible in 10 minutes averaged data.

Increasing the energy yield per unit area of Photovoltaic (PV) modules is one of the main challenges the PV technology is currently facing. In search of ways to address this issue, solar tracking systems are a favorable solution, which according to literature can enhance the energy yield by up to 45%. Another way is to opt for high-efficiency solar cells. Tandem cells have emerged as a promising technology, achieving efficiencies of over 30%. The integration of these tandem cells with sun-tracking techniques suggests high-performing solar modules. The present work develops a solar tracking model to simulate the performance of tracking PV systems equipped with tandem modules. The model will be incorporated into the PVMD toolbox, a PV modeling software developed within the PVMD group. This software can predict the energy yield of PV systems using self-consistent models for each aspect of the energy conversion.

The current version of the toolbox makes it impractical to include solar tracking due to the time-consuming nature of ray tracing used to compute the irradiance. Ray tracing generates sensitivity values that illustrate how sensitive is the module to incoming irradiance from any direction in the skydome. Initially, this work focuses on substituting ray tracing with an alternative faster approach to express sensitivity based on view factors. The view factor and ray tracing method are compared with respect to computational time and extent of agreement. It was found that the view factor can significantly reduce the computational time from over 12 minutes, as required in ray tracing, to a few milliseconds for a single module orientation. Additionally, the view factor method generates sensitivity values closely matching those from ray tracing. For instance, a mean RMSE of 1.2% between the two methods is achieved, for an albedo of 0.2 and module tilt of 30 degrees. Sun tracking aims to locate the module orientation that maximizes the in-plane irradiance. Directly calculating the irradiance for every orientation to identify the optimal, is not a viable option, as it requires substantial
time. Thus, sun tracking was expressed as an optimization problem and algorithms were employed to address it. Based on the prevailing sky conditions three optimization case studies were defined on an hourly basis: sunny, cloudy, and intermediate hours. Multiple algorithms were compared across the three cases with selected criteria the convergence to the optimum and runtime. Matlab’s surrogate solver and an author-developed algorithm were selected, as a satisfying solution, compromising those two criteria.

Finally, energy yield simulations were performed on perovskite-silicon tandem modules mounted on a dual-axis tracking system. Four locations were selected, representing different real-world conditions: Stockholm, Athens, Bombay and Bogota. Results show the module’s tilt dynamic adaptability to sky conditions: increas- ing nearly to the sun’s zenith when direct light dominates, and lowering when diffuse light is prevalent. Furthermore, the seasonal fluctuations of the energy gain of tracking systems are explored, with locations further from the equator such as Stockholm exhibiting the highest variability of 19% in winter to 36.9% in summer. In addition, the annual energy gained among the locations was found to span between 24.8% (Bogota) and 34.1% (Bombay). An important finding is the direct proportionality in gains from absorbed irradiance to DC and AC yields, illustrating a 1:1:1 ratio. Then, the effect of tracking technology on mismatch losses of tandem modules was examined. Results indicated that tracking has little impact on both the current and power mismatch. For example, the power mismatch losses slightly increased from 1.10-1.46% in static PV systems to 1.29-1.77% for tracking topologies in the locations examined. Moreover, the tandem’s annual energy gain is compared to silicon heterojunction modules. The analysis showed similar gains across locations for both cell technologies.

The increasing adoption of renewable energy sources, particularly photovoltaic (PV) systems in residential sectors has raised important energy balancing challenges due to the intermittent nature of energy generation. To address these challenges and prioritize cost savings for residential consumers, this research investigates the integration of battery energy storage systems (BESS) and dynamic pricing strategies through an intelligent energy management system (EMS). Given the stochastic nature of PV generation, market prices, and load profile it is still challenging to achieve optimal control. Therefore Reinforcement Learning (RL)-based EMS is proposed in this research to make real-time optimal control decisions. RL is a machine learning approach where an agent learns to make decisions by interacting with an environment to maximize cumulative rewards. In this study, a deep deterministic policy gradient (DDPG) RL architecture is chosen due to its capability to handle continuous action spaces. In addition, deep learning-based models are employed to forecast uncontrollable load, PV generation and market prices for the integration into the EMS for which Bi-directional LSTM (Long Short Term Memory) was found to be the most accurate for all three uncertain variables. The DDPG algorithm is trained with data from a single household from the Lucerne region, Switzerland for 30 days and tested for a week. The results showed that compared to a deterministic rule-based approach the RL-based EMS increased cost savings for the end consumer by 14.2% but reduced the benefits for the grid operator to alleviate grid congestion quantified in terms of load factor, peak power consumption and ramping. Further work could be undertaken in testing the model on more extensive data and finding the best trade-off between customer and grid operator benefits.

Renewable energy generation projects are often measured by their peak capacity. A wind farm rated at 25 MW will generate 25 MW of power under the right circumstances. This peak capacity is reached very little in practice. However, these generators are forced to purchase grid operation infrastructure that can handle this peak generation event. The high voltage grid connections are expensive and increasingly difficult to receive permits for. This work presents a solution in which the high voltage grid connection is undersized in comparison to the renewable energy generator. A battery energy storage system is installed in the local grid to solve the issue of excess energy generation (congestion).

A simulation of the local network has been built that models a battery energy storage system (BESS), the network and uses data from a solar park. A case study in which a 19 MW solar park is connected to the high voltage grid with a transformer of only 14 MW as well as a 14 MW | 30 MWh BESS on the network is investigated in the rest of the work. Furthermore a BESS control strategy for price arbitrage on the TenneT imbalance market is presented and encoded such that it can be optimised.

Four heuristics are presented that time and size the congestion issue in a manner the control strategy of the BESS can prepare for and solve congestion when necessary. These heuristics are tested against strategies optimized for revenue maximisation through price arbitrage. While the most aggressive strategies did not solve all the congestion events in these simulations, we found that the heuristic that takes the average generation of the solar park into account performs the best while remaining appropriately conservative.

A basic evolutionary algorithm is presented that optimizes a BESS control strategy for price arbitrage when the BESS is not needed on the local network to solve congestion. Although the strategies earn ~33% less revenue due to the congestion related limitations, the optimisation surrounding congestion does improve revenue by 2.58%.

The results presented in this work suggest that this setup of a local grid can be economically viable and that the BESS can solve the congestion issue when steered with an appropriate control strategy. We hope to inspire parties that battery energy storage systems can earn substantial revenue aside from solving issues on a (local) grid.

As the world is currently actively trying to reduce the consumption of fossil fuels, large investments are done in renewable energy sources and ways are sought after to electrify fossil fuel-intensive sectors. In line with these developments, the number of electric vehicles requiring access to the electric power grid has exploded putting increased pressure on the grid. One way to decrease the congestion in the grid is to make use of smart charging schedules for electric vehicles, with the objective of reducing peak demand and preventing the overloading of cables and transformers while reducing the cost of charging for electric vehicle owners.

The recent increase in the availability of real-life data has allowed the in-depth study of smart charging dynamics on a large scale through modeling and simulation. Mathematical optimization is a method that is often used to generate smart charging plans in state-of-the-art smart charging applications. However, while mathematical optimization can be very effective, as the objective function expands and more parameters are taken into account, the optimization becomes more complex and therefore require faster and smarter optimization algorithms. In addition, the recent availability of a vast amount of real-life data sets has made efficient data handling more important.

Machine learning is known to be an effective way of introducing Artificial Intelligence to smart charging algorithms. The use of reinforcement learning algorithms, a subset of machine learning could help overcome the disadvantages of mathematical optimization as trained algorithms are generally fast when predicting outcomes and have the potential to be accurate at the same time.

Therefore, the purpose of this work is to research the feasibility and additional benefit of machine learning to mathematical optimization-based smart charging algorithms. This is done through the development of end-to-end Q-learning and Double Deep Q Network reinforcement learning smart charging algorithms. The performance of both algorithms is evaluated on three separate case studies as well as on multiple different random cases and is compared to the charging performance of the average rate charging and mixed-integer programming algorithm benchmarks.

As a result, it becomes clear that for individual case studies the Q-learning and Double Deep Q Network agent are able to find cheap charging moments while charging the vehicle to 100\% battery capacity without violating charging constraints. However, when testing the performance of the Q-learning and Double Deep Q Network agents it becomes clear that the average charging performance is significantly worse than using the method of mixed-integer programming as the algorithms do not learn to generalize well.

Finally, the advantages and disadvantages of replacing mixed-integer programming with reinforcement learning are discussed as well as some limitations and recommendations for future work and improvement are given.

Given the urgency to shift to a cleaner energy mix, offshore wind energy is bound to be quickly deployed on a large scale in the next years. In order to meet the Paris Agreement, the current role-out rate of wind energy in the North Sea, that is 2 GW/year, needs to be increased to an average value of 7 GW/year over the period 2023-2040. This rapid increase of the share of uncontrollable, renewable sources in the energy mix raises a certain number of challenges, among which sustaining energy balance and transient stability. To organise the large-scale deployment of offshore wind in a strategic manner, the concept of hub-and-spoke has been introduced. It envisages modular offshore renewable power generation clusters (a.k.a. offshore hubs) designed to be easily and safely operated and interconnected.

Dealing with large power fluctuations constitutes a major challenge for each offshore hub. The variability of power supply needs to be mitigated through the integration of storage solutions. Offshore power-to-gas conversion for storage purposes has been widely investigated in the literature, under techno-economic or power balance points of view. However, detailed analysis of the effective connection of large scale electrolysers in offshore hubs are missing.

This thesis proposes a real-time simulation model of a 2 GW offshore hub integrating electrolysers in order to investigate their impact on the dynamic active power management of the hub. Two locations for the connection of the large-size electrolysers are studied: i) connection of an electrolyser to the AC common bus of the hub and ii) connection of electrolysers to the DC link of the back-to-back converter of each Type-4 wind turbine of the hub. An Electromagnetic Transient (EMT) simulation based analysis, necessary to capture the fast dynamics of the zero-inertia hub, is performed under different severe disturbances, such as a three phase fault and sudden large wind speed fluctuations.

In a second part, the thesis investigates the connection of the hub to the shore via a bipolar High-Voltage Direct Current (HVDC) link. It has been shown in the literature that bipolar links are more reliable than monopolar ones, as they can still be operated to transfer half of their nominal capacity in case of failure of one cable or converter. The work of this thesis is focused on developing and implementing a power sharing strategy between the two Modular Multi-level Converters (MMCs) constituting the offshore terminal of the link.

It was found in this thesis that both investigated locations for the connection of the electrolysers can be beneficial in different ways. Connected at the AC common bus, the electrolyser can provide centralized ancillary services, while an electrolyser connected locally at the wind turbine can consume the power generated during an islanding event. Regarding the power sharing in the HVDC link, two different strategies were developed, both enabling to unload one of the poles of the link. Further studies should be carried out to provide in-depth understanding of the potentialities of the two methods.

Solar energy is an abundant, scalable, and clean source of energy. With an exponential drop in prices of PV modules, more and more rooftop photovoltaic (PV) systems are being installed worldwide. Since these small-scale PV systems do not use expensive sensors, it is difficult to detect malfunctions for these systems. This could lead to lower energy generation along with financial losses for the owners. Thus, a method for PV yield monitoring is developed for early and remote fault detection. This method does not use the conventional analytical approach as it depends on inaccurately extrapolated weather data. Instead, the proposed method uses data from similar or neighbouring or peer PV systems for estimating the expected energy generation. By comparing the expected energy generation with actual energy generation, a faulty system can be flagged. In this project, information from about 12000 PV systems is used, which includes system design information such as location, number of panels, panel orientation, etc. along with the historical daily energy generation for periods ranging from two months to up to seven years per system.

In this thesis, a machine learning model was developed for predicting energy yields, which uses a Genetic Algorithm (GA) for optimization. This model splits the available data into system design, system location and system yield data. Thus, the model uses these as criteria for finding PV systems similar to the monitored system. Once good peer systems are located, system yield data of those systems are used for estimating the expected energy yields of the monitored system. The three criteria used by the model do not have equal influence on finding good peers, thus, the model had to be trained or optimization was done using the training data. Post optimization, the relative influence of system design: system yield: system location was found to be 0.125:0.875:0 with on average 16 good peers needed for accurate predictions. The proposed model has a mean normalized RMSE of 0.057 and about 95% of the systems tested had an R2 score higher than 0.85. The existing commercial software at Solar Monkey has a mean normalized RMSE of 0.082 and about 83% of the systems tested had an R2 score higher than 0.85.

The predicted energy generation calculated by the proposed model is compared with the actual energy generation to detect any malfunctions that may have occurred in the monitored system. Thus, 120 randomly chosen PV systems were analysed for faults. Based on this, a semi-automatic categorization framework was created with the proposed model as one of the criteria to detect common faults in the system such as missing data, under-performance, over-performance and false positives. Using the categorization framework, certain PV systems were found as interesting examples for under-performance with broken panels or string, over-performance with system size change and false positives. The model is especially useful for separating system design mismatch from actual system malfunctions. With the framework, it was shown how the proposed peer-to-peer model can be used for fault detection along with certain other models.

Since the introduction of SF6 in the 1950s, gas-insulated high voltage circuit breakers and Gas-insulated switchgear (GIS) have improved considerably, in particular concerning required drive energy for operation, compactness, and reliability. Nevertheless, high voltage insulation design has become increasingly challenging in recent years. Customer demands for reducing the physical footprint of HV equipment has led to an increase in operational field stress and therefore, much higher pressure on tolerances and increased susceptibility to defects. Dielectric design is based on how electrostatic fields are distributed and how much stress they can generate. In the course of conducting the intensive theoretical and experimental investigations on the dielectric design of insulation systems applied to high voltage power and pulsed power applications, it is becoming necessary to consider the influence of phenomena that have not been considered before. On top of that, as SF6 is one of the greenhouse gases listed in the Kyoto Protocol, SF6 usage regulations have been implemented in many industries. In Europe, SF6 is regulated under the F-Gas directive, which bans
or restricts its use for several purposes. Several studies have indicated that CO2 is a viable alternative to SF6 for the transmission and distribution of electricity.

In this context, this project aims to investigate a potential new phenomenon, the
occurrence of secondary discharges caused by propagating streamer or leader discharges in HV gas insulation. Learning more about such phenomena can help us improve the dielectric design or perhaps explain the occurrence of breakdowns reported in high voltage equipment for which no obvious cause could be found. Specifically, CO2/O2 mixtures will be studied, and its results compared to SF6. Breakdown in gaseous insulation is caused by propagating discharges that are external in ambient air, e.g. on a bushing, or inside high-pressure insulation, e.g. across the surface of a GIS insulator. During the propagation of such discharges and flashovers (streamer or leader), transient electric field enhancement may occur, leading the local fields to exceed the inception fields at other locations. This can result in secondary discharges.

Dielectric experiments using Image Intensifier and Photomultiplier tubes (PMT)
have been conducted at three different pressure ranges; SF6 and CO2/O2 were selected as the insulation mediums. A negative polarity electric field is expected to trigger the secondary discharge. An analysis of images obtained from optical investigations and time lag records obtained from PMT signals suggested that secondary discharges can occur at low pressures (0.2MPa) in both SF6 and CO2/O2. At higher pressures (0.4 to 0.6MPa), no secondary discharges were detected. The reason for this is that, at higher pressures, the breakdown field is higher leading to a faster propagation of discharge across that gap. Hence, the formative time lag of the discharge is very short (some 100푛푠 or less) and this short time is not sufficient for secondary discharge inception.