The shift towards zero-emission transport has driven rapid adoption of Heavy Goods Electric Vehicles (HGEVs) across Europe. This trend is also evident in the Netherlands, where their numbers have grown exponentially over the past six years. The integration of HGEVs into fleet operations introduces new challenges for fleet planning due to limited battery ranges and significant operational uncertainties in both the time and energy domains.

To address these challenges, this thesis introduces CertERoute, an Adaptive Robust Optimization (ARO) framework that allows for joint optimization of routing and charging scheduling under time and energy consumption uncertainty. It incorporates both depot and en-route charging while accounting for charger availability and other realistic operational constraints. Time-related uncertainties in service, waiting, travel, and charging durations are modelled using uncertainty sets, which require minimal assumptions about the underlying probability distributions. Energy consumption is modelled as a function of time-domain uncertainty, with environmental and vehicle-specific energy uncertainty factors accounted for through Monte Carlo Simulation (MCS). The adaptive design of the framework supports a two-stage decision process where routing and charger visits are planned in advance to hedge against worst-case scenarios, while charging amount and timing decisions remain flexible during route execution to avoid overly conservative and costly solutions.

To ensure computational tractability, the framework employs a Column and Constraint Generation (CCG) approximation method enhanced by a novel One-Step Look-ahead Pessimization (OSLP) algorithm, which selectively integrates only provably infeasible scenarios into the optimization problem. Despite its theoretical vulnerability to premature convergence, this algorithm empirically produces highly robust solutions. To improve scalability to larger instances, a multi-scenario Adaptive Large Neighbourhood Search (ALNS) metaheuristic is developed, integrating charging and timing decisions into neighbourhood generation to enable more informed solution exploration.

The framework is evaluated using representative yet synthetic European HGEV planning scenarios. The results highlight the scalability of the proposed solution methods and demonstrate that the resulting plans are highly robust under operational uncertainty. As a result, this work offers a practical foundation for integrated fleet and energy management systems and supports Shell eMobility’s strategic goal of enabling sustainable, cost-efficient transport through offering intelligent charging solutions.

This study investigates the integration of Vehicle-to-Grid (V2G) technology with Electric Car Sharing Systems (ECSS) to enhance their profitability while implementing active peak reduction strategies. Focusing on a station-based ECSS with a fleet of 24 electric vehicles (EVs) across five stations, the research develops a mathematical model to maximize system profits by optimizing driving, relocation, charging, and discharging activities enabled by V2G functionality.

The analysis encompasses scenarios with and without V2G under varying electricity pricing schemes - uniform time-varying and location-dependent time-varying - and seasonal conditions, including summer and winter. Results indicate that V2G significantly increases profits by selling energy back to the grid during periods of low driving demand and high electricity prices, especially in cases of location-specific price variations. Active peak reduction measures effectively lower peak load demand with minimal impact on profitability. However, the inclusion of V2G introduces complexities, such as higher vehicle relocation costs and increased battery cycling, necessitating advanced management strategies.

The findings emphasize the critical role of dynamic electricity pricing and regulatory incentives in maximizing V2G benefits, highlighting the potential for ECSS to support grid stability while reducing peak load demand. Future research should address real-world challenges, including user demand variability, vehicle availability, and battery degradation, to refine the model’s practical applicability. By combining car-sharing with V2G, this study contributes to sustainable urban mobility and renewable energy integration, offering a promising direction for innovation in transportation and energy systems.

Industrial electrification has a crucial role to play in reducing carbon dioxide emissions, and ensuring power reliability is an important factor in this process. Therefore, this thesis explores cost-effective storage to ensure the reliable operation of the renewable energy-based power supply of a 500 MW electrified industrial load in the 2035 Botlek area in the Rotterdam harbour. The study focuses on integrating electrical storage solutions to mitigate external power system risks that threaten reliability.

The research identifies key external risks to be energy shortages, power outages, frequency fluctuations and voltage instability. After a comprehensive review of current and emerging storage technologies, three storage mitigation strategies are formulated based on financial, technical, practical and other aspects. A synthetic industrial model consisting of generic component representations in DIgSILENT PowerFactory 2024 SP1 based on an electrified version of the Shell Pernis refinery is used. It is adapted to represent the Botlek 2035 study case, the storage assets are integrated and inputs are defined to represent on-site wind power generation and connection to the external future Dutch power system. 20% of peak demand is assumed to be flexible based on the residual load in the Dutch power system in 2035.
Via a dispatch simulation, the performance of each strategy is assessed. The loss of load expectation and expected energy not supplied are used as reliability metrics. The costs of plant interruption, storage system costs and electricity costs are used to select a strategy. Results are subjected to several input and parametric sensitivities.

Results indicate the combination of a Lithium iron phosphate battery, an alkaline electrolyser, magnesium hydride storage and a hydrogen turbine can ensure reliable operation most cost-effectively. Hydride storage is identified as a technology with much potential for industrial integration due to the advantages of waste heat utilization for hydride dehydrogenation. Extreme weather scenarios are advised to be covered with externally sourced fuel instead of on-site storage reserves. A 10.5 GWh or 315 tonne hydrogen storage system is expected to be required for reliable operation in the Dutch power system, when 20% of peak demand is modelled as flexible. This configuration yields to an average levelized cost of energy of 45.3 €/MWh during the year compared to 45.7 €/MWh without storage while also ensuring reliable operation without significant interruption costs.
This thesis provides insight into the integration of storage assets in electrified industrial networks in the context of the future Dutch energy system and lays a foundation for further research.

The growing integration of renewable energy sources presents significant challenges for the reliability of the energy infrastructures in The Netherlands. To address these challenges, dynamic electricity pricing has been proposed as a potential solution to align the electricity demand with the variable supply. While dynamic pricing can incentivize the shifting of flexible loads, its impact on the low-voltage grid remains under-investigated. This thesis explores the impact of dynamic pricing on grid congestion, especially in combination with EV charging and PV systems.


It is hypothesized that while dynamic pricing can effectively shift electricity consumption, it may also lead to concentrated periods of high demand. When consumers synchronize their electricity usage during low-price times, grid congestion can occur.
The study creates a simulation model using Python and PowerFactory to analyse the effects of dynamic pricing on the low-voltage grid. The simulations are based on real-world data from 2023, making use of hourly intervals of loads, EV charging profiles, and solar energy generation across three representative weeks. The model evaluates the voltage levels and power flows to assess the potential for grid congestion under various scenarios of EV and PV adoption.

The results indicate that dynamic pricing can lead to concentrated EV-charging loads, resulting in voltage dips and power spikes. These findings show that the current low-voltage grid may not be equipped to handle the increased demand and variability caused by dynamic pricing schemes. Thus, while dynamic pricing can offer a promising tool for demand-side management, additional measures are necessary to ensure grid stability and to avoid future congestion issues.

In conclusion, the study underscores the dual nature of dynamic pricing as both a valuable tool and a potential risk for the low-voltage grid. Both the voltage and power limits of the grid were found to be causes of congestion, highlighting the need to investigate both. Future research should focus on optimizing dynamic pricing schemes and developing robust grid infrastructure to fully leverage the benefits of renewable energy integration while ensuring grid reliability and resilience.

As the power system grows more complex and active, equivalent models have become a solution for modelling parts of the network that have limited observability or are confidential or too complex to simulate otherwise. In the past decade, this topic has also made its way to distribution networks because of its transition towards an active network, which was not the case before. The current grey- and black-box techniques for equivalent modelling create models that are fitted to the average dynamic response of the system or have limited applicability. Also, the existing methods lack extensive verification under different system conditions, and these works rarely focus on active distribution networks (ADNs) with multiple points of common coupling (PCCs). This thesis aims to develop an equivalent model that can estimate the dynamic response of an active distribution network with multiple PCCs and under diverse operating conditions and topological changes.

The proposed approach is based on a graph-time convolutional neural network (GTCNN) that relates available PMU measurements inside the distribution network on a graph structure 𝒢_GTCNN. The graph 𝒢_GTCNN is obtained by taking a modified line graph of the graph representation of the power system and is expanded using the Cartesian product graph rule to include the temporal dependencies of nodes on their past values. The inputs of the equivalent model are the voltage magnitude |V| and angle θ at the PCC and the initial power injections P₀ and Q at non-PCC nodes, while the model outputs the active P and reactive Q power at the PCC nodes. The GTCNN explicitly considers the initial power injections P₀ and Q₀ at non-PCC nodes to help the model learn how different operating conditions and topological changes impact the dynamic response. The DSO trains the equivalent model using simulation data or collected PMU measurements. The model is exchanged with the TSO every month, who can use the equivalent in co-simulation with their transmission network model to perform transient stability studies.

The GTCNN-based equivalent model showed promising performance as an equivalent model for transient stability. The GTCNN was benchmarked against two state-of-the-art Long Short-Term Memory (LSTM)-based equivalent models and a hybrid GTCNN-LSTM model. The evaluation was performed on a real Dutch distribution network using three datasets, each focussing on a different system condition: different fault events, different operating conditions and different hidden topological changes. The GTCNN-based equivalent model had a mean-squared error (MSE) below 0.02 for each dataset, which means it can accurately reproduce the dynamics. This accuracy is comparable to the LSTM-based equivalent models, but the GTCNN could train 4x faster. The GTCNN also showed good generalisation performance, as its accuracy did not decrease on the validation and test sets. A study on scaling performance suggested that the MSE of the GTCNN-based equivalent model increases slower than that of the LSTM-based models while its training time increases faster. Therefore, the GTCNN-based equivalent model trains faster for smaller ADNs but will be more accurate with more measurement nodes. However, the proposed GTCNN has difficulty learning the response at different close-by PCC terminals if the dynamics are different.

The developed GTCNN-based equivalent model can predict the dynamic response accurately under changing topologies and operating conditions at a similar performance level to existing LSTM-based approaches. However, its training time is much faster, which can result in a more accurate equivalent model by a more frequent model exchange between the DSO and TSO or a more extensive dataset being used to train the model. In future research, the GTCNN performance will be evaluated on a more comprehensive dataset containing all three system conditions to establish how much data is needed to train the equivalent model accurately. Also, the system frequency will be considered as an additional input. Moreover, its scaling performance will be evaluated more extensively and with a more efficient coding implementation. Furthermore, a heterogenous graph convolutional operator will be implemented to learn the connection per relational type (source node - edge type - target node). Finally, the co-simulation interface between the equivalent model and popular simulation tools will be explored.

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 investigates the performance of a novel building-integrated photovoltaic (BIPV) window solution equipped with bifacial photovoltaic (bPV) cells and reflective Venetian blinds through a PV modelling approach. By modelling this PV window solution in BIGEYE, the study aims to evaluate how different types of blinds and tilt angles can help to boost the overall performance of the bPV window and to quantify the potential energy gains under varying weather conditions. First, the optical model was validated against experimental data to ensure it could accurately simulate the irradiance received at the front side of the window plane. The optical properties of the Venetian blinds were also defined to quantify the irradiance gain at the rear side of the bPV window. Then, the electrical model was also validated to make sure the model could accurately simulate the electrical behavior of the bPV window demonstrators in real operational conditions. The validated models were then used to test the effect of using different blind configurations in the bPV window’s performance.

In this way, three different types of Venetian blinds were tested: S157 (grey colored), S102 (white colored) and V95 (silver colored). The analysis revealed that, when deployed at a tilt angle of 60°, these types of blinds can help to boost the irradiance gain at the rear side of the bPV window by up to 16%, 20% and 25%, respectively, on a fully sunny day. The study also demonstrated that blinds positioned at 60° consistently achieved the greatest absolute boost in temperature-corrected performance ratio (TCPR) of the bPV window over a typical meteorological year (TMY) in Eindhoven, The Netherlands. In this way, the V95 blind type at a tilt angle of 60° demonstrated to be the optimal blind configuration for boosting the performance of this bPV window. Furthermore, an in-depth analysis was conducted to examine the influence of additional factors on the performance of this bPV window, such as the angle of incidence (AOI) between the sun and the blinds' surface normal, which showed to have an inverse correlation with the bPV window’s TCPR, highlighting that the performance of the window tends to decrease as this AOI increases.
Furthermore, the study quantified the Yearly Specific Yield (YSY) boost achievable with the optimal blind configuration, finding that a maximum boost of 19% can be attained on a TMY in Eindhoven, 18% in Stockholm, and 17% in Valencia. The difference in performance across locations is attributed to variations in meteorological and geographical factors, such as diffuse fraction and AOI. Overall, the findings provide valuable insights into how different blind configurations can enhance the performance of this bPV window and contribute to enhance the energy performance of high-rise residential towers and office buildings.

The use of DC microgrids in combination with photovoltaic (PV) energy generation sources results in higher efficiencies by avoiding AC to DC conversion. Moreover, bipolar DC grids allow for more system flexibility and increased reliability at the cost of an additional conductor when compared to unipolar DC. However, maximum power point tracking (MPPT) converters are still required in order to optimize the power delivered to the microgrid by PV systems. The development of more efficient and cost-effective DC-DC MPPT converters will be proven vital considering the expected growth in solar PV installation as a consequence of the relentlessly increasing global energy demand.

The objective of this thesis is to design a DC-DC MPPT converter for bipolar DC microgrids based on gallium nitride (GaN) high-electron-mobility transistors (HEMTs). GaN transistors are wide-bandgap semiconductor devices that can operate at very high switching frequencies due to their superior electron mobility compared to other semiconductor materials used in electronic switching. Increasing the operational frequency of an electronic converter allows for smaller passive components, which in turn has the potential to result in system size and cost reductions. A complete electrical and thermal analysis of the performance of these components is simulated and presented in this report, and the implications of employing these devices for the selected application are evaluated. The design overview of the DC-DC MPPT converter using these GaN HEMTs is also showcased in this work, as well as the results obtained from testing the assembled DC-DC MPPT converter for bipolar DC microgrids. The main conclusions of this work aim to clarify the advantages of GaN devices in MPPT applications, as well as demonstrate a design solution to the integration of MPPT converters to DC microgrids. The project was developed in collaboration with the company DC Opportunities, and the designed converter was assembled and tested in their laboratory.

The EU strives to lower greenhouse gas emissions. To reach this goal, many energy intensive processes in the residential sector such as heating and transportation will be electrified using heat pumps and electric vehicles (EVs) respectively. Simultaneously, a transition of electricity generation to sustainable sources will take place, necessitating an increased adoption of rooftop photovoltaic (PV) systems.

The adoption of PV systems, heat pumps and EVs, also known as low carbon technologies (LCTs), can increase three-phase unbalance in low voltage (LV) distribution networks as many of these components will be connected to a single phase of the three-phase network. Threephase unbalance is undesirable in a three-phase system, as it causes among others, energy losses and a suboptimal use of network capacity.

The aim of this thesis is to evaluate the impact of different combinations and penetration levels of LCTs on three-phase unbalance in different real LV distribution networks through simulations and how unbalance is affected by LCT location, season and LCT control schemes.

Simulations were performed on six different grids, varying in level of urbanization and loading, with increasing levels of LCT penetration (0%, 50%, 80%, 100%). In these simulations, LCTs were integrated in varying combinations (PV & EV, PV & HP and PV & EV & HP). For every simulation, the maximum and mean voltage unbalance factor (VUF) was determined. Seasonal effects and the effect of an LCT control scheme were also evaluated.

Simulations showed that the voltage unbalance factor exceeded the legal limit of 3% for two of the six grids for high levels of LCT penetration when all LCTs are integrated. Combining all three LCTs resulted in the highest unbalance levels. Varying the locations of the LCTs resulted in significant differences in unbalance levels. Comparing a winter week with a summer week, the overall unbalance is similar, however, the contribution of the PV systems to the unbalance is increased, while the contribution of EV chargers and heat pumps is decreased.
The effect of the LCT control scheme was limited.

As the integration of LCTs will increase considerably in the near future, three-phase unbalance levels exceeding the limit of 3% will occur more often. To prevent unbalance levels from structurally exceeding the legal limit of 3%, more effective control schemes should be designed and implemented.

Airports and airlines are examining and committing to the electrification of Ground Support Equipment (GSE). To be able to estimate the required quantity of eGSE, the charging requirements of eGSE, the change of airport electricity requirements, and the scheduling possibilities of eGSE charging for the existing turnaround procedures, a model was developed to simulate and optimize the GSE operations at airports. This was done by means of a Task Scheduling Problem (TSP), that is optimized using Mixed-Integer Linear Programming (MILP). A case study was performed on KLM's GSE fleet at Amsterdam Airport Schiphol. Based on this, it was concluded that there is no difference in the capacity that can be achieved for GSE types that can last an entire day on a single battery charge. However, another group of GSE types experiences battery depletion before the day concludes, requiring measures to maintain the capacity. The results indicate the model's suitability for strategic decision-making. Next to that, the model is effective on an operational level. The use of the model has the potential to make the use of resources in the operation more efficient.

With the development of sustainable energy technology and the increasing emphasis on environmental protection, the popularity of electric vehicles is increasing. However, As the number of electric vehicles increases, so does the pressure on the grid due to charging. If a large number of electric vehicles are charged at the same time, the grid will inevitably be overloaded. For this problem, smart charging can be a good solution. The smart charging algorithm determines when it is optimal to charge the electric vehicles(EVs) based on factors like Photovoltaic(PV) generation, local load consumption, and energy price. For customers, this means the total cost of charging is the lowest, for the power system it means charging when the network is not overloaded. Furthermore, Smart charging can reduce the need for grid infrastructure investments and reduce energy costs. From an energy-friendly and economical perspective, this is worth doing.

Up to now, there is a large amount of literature study about smart charging. However, most of the literature is based on pure software analysis rather than hardware experiments. In this thesis, a hardware-in-the-loop-based simulation is presented. The basic idea of hardware-in-the-loop simulation is to include a part of the real hardware in the simulation loop. The advantages of hardware-in-the-loop simulation are obvious. Compared with pure software simulation, hardware-in-the-loop(HIL) simulation has the real response of the hardware components, and it is not very complicated to set everything up. In this thesis, the testing and the evaluation of the system are carried out in real-time because the HIL setup includes an OPAL-RT OP4510 DRTS. The simulation of the distribution grid is running on the OPAL target by the software ePHASORsim. The rest part of the testbed consists of a Programmable Logic Controller (PLC) and two bidirectional Direct Current(DC) power supplies to form an Alternating Current(AC) source and an Alfen charger. The smart charging algorithm is written in Python code. By calling Gurobi, smart charging algorithms can calculate and decide the optimal current setpoints for each EV connected to the distribution grid in real-time.

With the integration of solar panels in urban areas, where shading is a common issue, it is desired to realize more shade-resilient designs. A DC-DC boost converter is an electronic device that can be used for maximum power point tracking (MPPT). When implementing MPPT at sub-module level, the shade resilience of the module can be increased. This thesis aims to investigate the feasibility of implementing planar inductors into c-Si solar cells using the numerical simulation software COMSOL Multiphysics®. Through simulations, the inductance and resistance values for various planar coil geometries were obtained. Subsequently, the feasibility of using these coils in a sub-module DC-DC boost converter was investigated. For this study, a specific case was assumed where a Gen 2 Cell - 160 mm c-Si cell is used with a DC-DC boost converter with a duty cycle of 0.5 and a current ripple of 20%. Given the parameters from these devices, the minimum required inductance of 1.20 μH for a frequency of 200 kHz and 2.40 μH for a frequency of 100 kHz were obtained. The series resistances that can be added to the solar cell while still maintaining high efficiency, range from 1.0 mΩ to 4.8 mΩ, for a 1% to a 5% power loss, respectively. This is all for single-cell power conversion. Two different approaches have been examined; the screen-printed coil(s) approach and the screen printing thickness exceeding coil(s) approach. The former studies coils that do not exceed the maximum thickness achievable with screen printing. As such, these coils could be simpler to implement into a solar cell production line than coils with a higher thickness. In this research, for thicknesses below 300 μm a best-case scenario is studied where the skin effect is neglected. The latter studies a larger range of thicknesses, not taking any specific production method into account, but the skin and proximity effect is taken into account. The screen-printed approach is limited to a thickness range of 10 μm to 200 μm, thus simulated by the 2D model. For the air-core single-cell inductors for DC-DC boost converter applications, the inductance and resistance values were either below the 1.20 μH threshold or above the 4.8 mΩ threshold, respectively. These thresholds change depending on the number of series-connected cells. The inductance of a planar inductor can be boosted up to 200% when adding a magnetic material in a sandwich structure, however, this will also increase the costs of the application. In a boosted-inductance case, the 3-turn coil with a thickness of 200 μm, and a 5% power loss is the only configuration that is feasible for a 200 kHz DC-DC boost converter application, with an inductance of 1.569 μH and a resistance of 3.45 mΩ. By increasing the number of coils per cell, the power dissipation in the coils can be reduced. Using this approach, six feasible topologies can be created. One of those is the nine parallel-connected 8-turn coils with a thickness of 200 μm, reaching an inductance of 1.210 μH and a resistance of 3.186 mΩ, allowing for a 200 kHz DC-DC boost converter application without any magnetic inductance increase. All six options are feasible for integration on a single cell using screen printing, however, the actual resistances will be higher due to the skin and proximity effect, and single-cell integration will be too expensive for industrial applications. When studying thicknesses above the screen printing limit, only the 4-turn coil with a thickness of 300 μm, a magnetically increased inductance of 2.718 μH, and a resistance of 4.01 mΩ, is feasible for the 100 and 200 kHz frequencies. This case however does neglect the skin effect. For the 3D model and thicknesses above 400 μm, the skin effect significantly impacts the resistance. When series connecting multiple cells to a single coil, the only feasible option is the eight-cell connected 8-turn coil for a thickness of 200 μm and 300 μm, both reaching a magnetically increased inductance of 10.431 μH and resistance of 23.33 mΩ and 15.56 mΩ, respectively.

With the current urge around the globe for switching to “green energy” and Renewable Energy Sources (RES) utilization, the electrification of the transportation sector has become a necessity. While the emerging Electric Vehicles (EVs) can provide several benefits to the power grid, such as voltage and frequency regulation and power quality improvement, the large and immense EV fleets penetration can have the exact opposite results, if no suitable charging method is utilized. Therefore, the uncontrollable EV charging, that is being utilized today, will not be capable of offering sustainable energy to the future EVs with respect to the power grid limits and hence, today’s research field has turned to optimized smart-charging techniques. However, most of these algorithms treat the problem of smart-charging deterministically, assuming 100% accurate prediction of the input data, while it has been shown that the optimality of results can be seriously deteriorated even under small prediction error.This thesis purpose is to address the impact and potential management of several uncertainties related with EV smart charging: PV Generation, Load Demand, arrival SOC, Arrival and Departure time of the EVs & (Frequency Containment Regulation) FCR Reserves provision uncertainties. The main handling technique, utilized in this thesis, is the Robust Optimization (RO) approach, taking advantage of its very lower computational expense and protection against the worst-case scenario of the uncertainty, with the combination of the Receding Horizon Optimization (RHO), that is already implemented in the Benchmark algorithm. After improving the Benchmark algorithm by providing it with prediction feature, this thesis proves the capability of improved robustly EV charging with the combination of RO-RHO approach with prediction, which reduces the economical (in terms of charging costs/income) and “customer satisfaction” (in terms of unfinished charging gap impacts) of all the uncertainties considered. A more robust and “realistic” FCR reserves provision model has been developed, as well. Last but not least, the well-known RO drawback of potential over-conservativeness, in other words high deterioration of optimality at expense of robustness, is address for every uncertainty.The results of the investigation provide valuable results about which uncertainty has the greater impact, is more robustly manageable or inflicts the highest “Price of Robustness”, regarding overconservativeness. For example, it has been found that the FCR reserves uncertainty inflicts the highest economical and charging gap impacts, however it is the most robustly manageable uncertainty as well, if RO and prediction feature are utilized. Moreover, while arrival SOC and Parking Time uncertainties affect highly the unfinished charging gap, the PV Generation and Load Demand have practically only economic impact, which is also lower than the other uncertainties, hence they are defined as the “uncertainties with the least impact”. Finally, there are 3 types of nodes studied: the “Home” Node, the “Semi-Public” Node & the “Public” Nodes. Observing their behavior during the different uncertainties’ study cases, interesting results have been found about their robust management and affection by the various uncertainties.

PV systems in urban environments frequently get shaded by nearby objects, which greatly reduces their yield. A potential solution for such situations is the use of reconfigurable PV modules. They are a type of modules which can reconfigure themselves under varying illumination conditions to ensure optimum yield at all times. The individual PV cells of the module are grouped into reconfigurable units/ cell blocks which can be interconnected in various ways to form a variety of unique configurations. The modules have an inbuilt algorithm which is responsible for controlling the reconfiguration process. There are various kinds of reconfiguration algorithms found in literature, each working with varying input parameters and operating principles. One such algorithm is the short circuit current sensing (SCCS) algorithm developed by the PVMD research group at TU Delft. The algorithm is meant for series-parallel
connected reconfigurable modules and uses each cell block’s short circuit current as input. It is a synchronous reconfiguration algorithm, which means it runs at regular time intervals and not when there is actually a change in the module’s irradiance conditions. Therefore, the aim of this thesis was to develop a robust reconfiguration algorithm with a shade detection mechanism. Consequently, numerous versions of a reconfiguration algorithm with shade detection mechanism were defined and tested using a simulation framework for modelling reconfigurable PV modules. It was observed that the reconfiguration algorithm that performed the best in terms of yield, reconfiguration count and accuracy was the one referred to as the ”CC4AP+IR algorithm”. The algorithm uses the module’s and a single cell block’s operating current and voltage as inputs for shade detection. It was observed that the algorithm reconfigured a fraction of the time that the SCCS algorithm reconfigured and had a DC yield comparable to the SCCS algorithm. The new algorithm was also tested using real-life data to validate its performance. This experiment indicated that if the module were to be run using the new reconfiguration algorithm instead of the SCCS algorithm, it would have run only 6% of the time that the SCCS algorithm ran, but have a comparable DC yield. The significantly lower reconfiguration count is relevant because in practice, the PV modules are connected to power converters which will not be able to tolerate the high fluctuations in output owing to frequent reconfigurations.

The energy transition is driving extensive changes to global energy systems, including the electrification of heating and mobility, increased renewable electricity generation, and a shift towards sustainable gases like green hydrogen. These changes will need to be accompanied by major investments in energy infrastructure, especially considering the goals of climate-neutrality envisioned for 2050. In integrated energy system planning, the system-level interactions between, for example, storage, electricity networks and gas networks are taken into account and considered as a whole. This thesis investigates the potential benefits of integrated energy system planning, focusing on the distribution grid level. The analysis is carried out for a case study in the Sterrenburg region of South Holland. Power-gas integration and the integration of electrical energy storage with batteries are investigated using hourly demand and generation profiles based on 2050 scenarios. An expansion planning model is developed to determine the optimal investments for the 2050 scenarios, starting from the existing network. This mixed-integer linear programming model uses PyPSA, an open-source energy system modelling toolbox. The optimisation objective is a combination of investment and operational costs. The results demonstrate spatial and economic benefits from including electrolysis, and limited effects from electric storage and gas-to-power within the scope of the case. There are many possibilities for broadening and deepening the scope in future works. For example, different combinations of integration such as heat networks and long-term gas storage could be considered together.

Electric vehicles (EVs) are going to take over the conventional cars industry like a wave in the foreseeable future. The exponential increase in sales of EVs is seen and market study has predicted that EV sales will increase by 15%–20% of total new vehicle sales by the year 2035. Despite being superior to fuel-driven vehicles, the increase in the number of EVs will have its challenges. For example, EVs are heavy uncontrolled charging loads that can lead to power surges in the grid which would then lessen the grid reliability. It is important to address the important question on how to maintain EV charging and use it in such a way that it will not pose a threat to the grid rather become its helping hand? The focus of this thesis will be to answer the aforementioned question. The main objective of the thesis is to minimize the cost of energy for charging EVs by developing and implementing a smart charging algorithm with V2G services considering the effects of battery degradation. The said objective is achieved by first formulating a mathematical model for the minimization optimization problem. Several assumptions were made to mimic real-life situations. Then an algorithm is developed and case studies on four sensitivity parameter are done. The parameters selected are the cost of penalty, cost of PV generation, cost of selling energy, and grid import power limitations. The case study is done to verify the sanity of the code and it has been observed that the algorithm developed prioritizes PV utilization, minimizes the overall cost of the node as well as EV charging cost. The algorithm incorporates the V2G application of EVs to support renewable (here, PV) sources. It has been observed that EVs that perform V2G generate financial benefit by exporting power to the grid and reduce the grid involvement by charging other EVs through V2G. V2G has adverse effects on the battery lifetime. Frequent charging and discharging can degrade the battery much quicker. The thesis tackles this problem by incorporating a battery degradation model in the developed algorithm for V2G. Two degradation model are used and compared to check the effectiveness and control while performing V2G. The first model is a simplified one and calculates the cost of battery degradation using energy exported from the battery. The second models the degradation considering one of the stress factors of cyclic aging which is C-rate. The newly developed model is observed to be more effective as the model is dependent on charging and discharging current rates. The EVs are seen charging just to their requested energy demand with relatively less power as compared to when the simplified degradation model is implemented. The V2G functionality is also observed to be reduced and utilized to reduce the grid involvement and reducing the overall cost of energy. Finally, the algorithm is compared with the uncontrolled charging algorithm. The developed smart charging algorithm is observed to reduce the charging cost of EVs significantly when the PV generation is sufficient. The smart charging algorithm utilizes the PV generation in supplying the energy demands of the local loads and EVs. This reduces the involvement of the grid. The grid is only observed to be involved when the cost of buying energy from the grid is lower to reduce any possible penalty for unfinished scenarios or when the PV generation is insufficient.

The growing global energy demand and correlated rise in carbon emissions is forcing us to increase the use of renewable sources. The residential sector represents a large part of the total energy consumption, and European governments are investing in distributed PV to increase the renewable share in this sector. However, on top of the solar power variability, residential systems are also characterized by very unstable load profiles. This issue can be solved by incorporating energy storage, that has many technical and economic benefits for the prosumer, especially if a long-term seasonal storage technology is used. Among all the available storage types, after an extensive literature study, some developing technologies proved to be suitable for this purpose: redox flow batteries (all-vanadium, hydrogen-bromide, zinc-bromide) and hydrogen systems.

For this work, a model was built in MATLAB Simulink to study and reproduce the behaviour of these storage systems in a grid-connected residential environment, and an optimization was set up to find the optimal sizing of the components and investigate the economic feasibility of the whole system. The models results proved that in the present scenario, storage integration is still too expensive with these technologies. However, future projections with different incentive scenarios demonstrate the potential of vanadium and zinc-bromine batteries, and highlight a dramatic cost reduction for hydrogen systems.

Multijunction devices based on thin film silicon are useful for a range of applications due to their low costs, better performance at higher temperatures, potential for flexibility, and light weight. One particularly interesting application of such devices is in solar-to-fuel conversion. In light of this, the goal of this thesis was to suggest ways to better design multijunction devices based on thin film silicon. Three main research objectives were identified.
The first objective was the development of a high VOC top junction. Cells based on a-Si and a-SiOX were optimized by varying gas flows during the deposition of the intrinsic absorber layer. A maximum VOC of around 890 mV was obtained from cells based on both absorbers. Based on these results, it was possible to identify unique advantages of using either a-Si or a-SiOX as the top junction. For the second research objective, solar cells based on a-SiGe were studied. The impact of varying the bandgap profile in a-SiGe absorbers on the device performance was investigated with a set of experiments. Subsequently, a set of semi-empirical equations were motivated and fitted to the experimental data. These relations could approximately show how bandgap profiling affects device performance in a more general way. The presented approach may be used to expedite the design of a-SiGe based cells for use in different types of multijunction devices. Finally, a systematic optimization of tunnel recombination junctions (TRJ) was conducted. Experiments were done to study the impact of using different TRJs in two types of tandem devices: a-Si/a-SiGe and a-SiGe/nc-Si. Different features of the TRJ were varied separately: p-doped layer(s) used, thicknesses of the p-doped layer(s), and the n-doped layer(s) used. The best performing p-doped layers was identified among the studied combinations. For the n-layer(s), the experimental data suggested the different roles played by n-a-Si, n-nc-SiOX, and n-nc-Si in the TRJ. This observation could be used to design an optimized combination of n-layers for the TRJ. Collectively, the work done as part of each research objective aims to contribute to the better design of multijunction devices based on thin film silicon alloys.

For the continued growth of PV capacity, it will be necessary to store vast amounts of energy to overcome seasonal differences in energy generation. To this end, the conversion of electricity to liquid or gas fuels could play an important role in the future energy system. There are numerous ways to create fuels from solar energy, the possibilities range from completely monolithically integrated photoelectrochemical (PEC) devices to configurations where the PV is completely decoupled from the electrochemical reaction. This thesis is part of the DISCO project, where the aim is to create multi-junction PV cells that can generate a high voltage and integrate these into PEC devices. The aim of this thesis project is to optimize the photovoltaic performance of a triple-junction device based on crystalline silicon (c-Si), nanocrystalline silicon (nc-Si:H) and amorphous silicon (a-Si:H).

The bottom junction of this triple-junction device is a wafer-based silicon heterojunction (SHJ) cell. The used wafer should have a smooth surface texture, since previous research has shown that growing a nc-Si:H absorber layer on the regular random pyramid surface texture results in the formation of cracks. To this end, three different proposed smooth surface textures were compared amongst each other. These surface textures were a smooth pyramid surface texture, a hexagonal surface texture created with photolithography, and a texture of crater like surface features obtained by etching a sacrificial layer. It was found that from these three textures, the best performing SHJ cells could be made from wafers with a smooth pyramid surface texture. With respect to the other surface textures, the smooth pyramid surface texturing approach leads to the smallest fraction of exposed <100> crystal orientations, which are more challenging to passivate. The performance of the SHJ cell is extremely sensitive to the passivation quality of the intrinsic amorphous silicon layers on both side of the wafer. By optimizing this passivation layer and the doped layers, SHJ bottom cell efficiencies of over 18% were achieved.

The middle junction of this triple-junction device is based on a nc-Si:H absorber layer. By varying the crystallinity of this i-nc-Si:H absorber layer, it was found that optimal photovoltaic performance is achieved for a crystalline volume fraction between 50% and 65%. Furthermore, the thickness of this absorber layer can be varied to improve the current matching between the different subcells in the triple-junction device. To this end, the effect of a varying absorber layer thickness on the single-junction PV performance was investigated. It was found that the absorber layer thickness is a trade-off between on the one hand Voc and FF, and on the other hand Jsc. For example, doubling the absorber layer thickness from 2.5 µm to 5.0 µm results in a Jsc increase from 18.2 mA/cm² to 20.8 mA/cm², whereas the Voc and the FF respectively decrease from 506 mV and 0.65 to 481 mV and 0.59.
Once the bottom and middle junctions were optimized, different material combinations in the tunnel recombination junctions of the multi-junction device were investigated. It was found that p-nc-SiO_x and n-nc-SiO_x with appropriate doping levels and contact layers were the best performing materials in both tunnel recombination junctions. When the Voc of the tandem cells was compared to the sum of the single-junction cells, the optimized tunnel recombination junctions resulted in a Voc loss of 7 mV in the top tandem (nc-Si:H/a-Si:H), and 51 mV in the bottom tandem (SHJ/nc-Si:H).
By making use of all the performed optimizations described above, a triple-junction device based on c-Si/nc-Si:H/a-Si:H was processed. The final device had a PV efficiency of 13.63%, with a FF of 0.755, a Voc of 1982 mV and a Jsc of 9.11 mA/cm². To the best of our knowledge, this is a world-record efficiency for this type of solar cell configuration. If it is managed in future research to match the current over the different subcells more evenly, this device has the potential to achieve a current density over 12 mA/cm².

With the strict Paris climate goals, all eyes are set on the development of sustainable solutions that can replace or improve commercial solutions. This thesis revolves around the design and implementation of a subsystem for the creation of a solar powered drone. The goal of the solar powered drone is increasing the flight range of a commercial drone with the addition of photovoltaic (PV) panels. The subsystem designed and implemented in this thesis is split up into two separate systems namely, the control part and the image processing part. The control part revolves around the design, implementation and validation of Maximum Power Point Tracking (MPPT) for the PV panels. The image processing part revolves around the design, implementation and validation of an albedo processing pipeline.
For the MPPT, a variable step size perturb and observe algorithm has been implemented with a worse case tracking delay of 40 milliseconds during a change of irradiation from $900 W/m^2$ to $100 W/m^2$ and and more than 99\% power efficiency in steady state conditions. For the albedo pipeline, multiple RAW images are loaded one after another. Vignette correction is applied making the pixel values at the edges and the center of the image compatible with each other. The user can annotate a known reference target that is used to calibrate the albedo generation. Finally, the images are stitched together using Open Drone Map to create a complete albedo map. The results indicate accurate visible band albedo generation, however additional sensors would be needed to obtain wide band albedo generation.