Bottom up load forecasting, is a technique where energy consumption forecasts are made on lower spatial levels, after which the resulting forecasts are aggregated to form forecasts of higher spatial levels. With the current move to renewable energy sources and the importance of reducing the strain on an already congested electricity grid, accurately forecasting both the location and time of future energy consumption has become more important than ever. To this end, this paper analyses the impact of applying bottom up load forecasting to different spatial levels on the electricity grid, including appliance, household, community and city spatial levels. Energy consumption data for these spatial levels were gathered from the 15-minute Texas and California datasets from Pecan Street Dataport. The results obtained in this study, suggest that energy consumption volatility at lower spacial levels and forecast difficulty at higher spacial levels, play an important role in the performance of applying the bottom up load forecasting technique to energy consumption forecasting problems.

Explainable artificial intelligence has in recent years allowed us to investigate how many machine learning methods are creating its predictions. This is especially useful in scenarios where the goal is not to predict a variable, but to explain what influences that variable. However, the methods that have been created thus far do not focus on specific domains and only give scientific relations between variables. In the case of a production process, there exist a multitude of related variables, constrained in different ways, thus requiring a more sophisticated method than those which are universally applicable. The target of this research is a bottle filling machine at a large carbonated drink manufacturer, of which a great amount of data has been stored. The product quality, in this case the amount of CO2 in the filled bottles, is subject to high amounts of fluctuation, leading to a lot of waste. To further contextualize the problem of bottle filling, an artificial dataset is created to serve as a basis for visualization and evaluation. Using literature on carbonated drink filling, multiple hypotheses were then formulated, independently from the thesis company, as a target for the evaluation process. To address the problem of determining what factors influence bottle CO2, a methodology is proposed which aims to provide explainable quality prediction of production process output, while using available domain knowledge to increase the plausibility of explanations. The methodology makes use of counterfactual explanations for tree ensembles and process state forecasting to show how a process state can be changed to yield better product quality. Evaluation on the artificial and real datasets show that the methodology is able to effectively predict the amount of CO2 in a bottle, while giving changes to the current process state which improve the product quality. It is also shown that the results produced by the methodology are in line with the formulated hypotheses. Bottle CO2 is influenced greatly by temperature changes in the filler and these temperature changes are caused by flow of cooled lemonade through the process. As these flows are dependent on the filling speed, irregular filling is suspected to be the root cause for the bottle CO2 fluctuations.

West IT Solutions is expanding into a new focus area, ”Smart Planning”. They want to develop new modules for the Enterprise Resource Planning software they resell, Odoo. Odoo allows companies to manage their projects, employees and other resources. The goal of this project was to create a software product that assists users in creating plannings for their projects in Odoo. The product should offer the following features: Visualise a planning for the user’s projects, Facilitate manually adapting the planning and Propose an automatically generated improvement of the planning. During the research phase of the project we met with a client of West, TNO. TNO explained they are performing research for the Dutch navy to create software that assists planners in creating schedules for battleships. With this case problem in mind, we created a list of requirements our software product should adhere to. We discovered that the case problem West provided us with is a subset of the case problem TNO provided. This meant that if we created a model for the case problem of TNO, the model would work for the case problem of West as well. After the research phase was completed, we started development on the software product. The development phase consisted of seven weeks, and in each week new functionalities were added to the product. In addition to adding new functionalities, we also kept refining our models to improve the performance of the automatic planning generation. The final software product consists of an interface that visualises a planning, a model that can be entered into external algorithms to automatically generate plannings and a module that allows users to import existing projects from Odoo. The user can view their planning through a visualisation that allows them to modify it using a drag and drop interaction. These modifications are setting preferences, like starting times for activities or the assignment of certain persons to certain activities. Once the user has set their preferences, they can ask the software to find an optimised planning that keeps their preferences in mind. At the end of the project, we gave a demo to TNO and they were impressed with the results of the project. West IT was also happy with the delivered product and is interested in further developing it

With the decision made to act upon climate change, the remaining question is: "How?". Economic theory suggests that the most efficient method is by means of market-based policies. These policies are often designed based on Integrated Assessment Models like DICE, which is the subject of this thesis. Integrating uncertainty into this model has been the aim of the thesis. The stochastic version of DICE, EICE, showed to be sensitive to the distribution of climate uncertainty, but results were less explicit then expected. Overall, the conclusion is that both modest and more strict mitigation policies can be justified based on EICE.

As traffic demands are ever increasing and building new infrastructure poses challenges in densely populated areas, it is important to optimally utilise existing infrastructure. Short-term traffic forecasting can help with this task, as its predictions can help to prevent congestion by rerouting vehicles. Recently, neural networks developed for traffic forecasting have lead to an unprecedented accuracy, but deploying these on large scale can be difficult as the resulting models likely overfit to the highway they were trained on. This thesis therefore performs an in-depth assessment of the accuracy of neural networks traffic speed predictions on highway stretches containing different types of congestion patterns. By relating the results back to traffic flow theory, these results can be put into perspective.

The European Union is on course for introducing a European Digital Identity that will be available to all EU citizens and businesses. This will have a huge impact on how citizens and businesses interact online. Big Tech companies currently dictate how digital identities are used. As a result they have amassed huge amounts of private user data. Movements like Self-Sovereign Identity aim to give users control over their online identity. TrustVault is the first data wallet that gives users back control not only of their identity, but all their data. TrustVault allows users to store all their data on their smartphone and control who they share it with. The user has fine-grained access control based on verifiable user attributes. EBSI connects TrustVault to the European Self-Sovereign Identity Framework allowing users to use Verifiable Credentials from public and private institutions in their access control policies. The system is serverless and has no Trusted Third Parties. TrustVault replaces the for-profit infrastructure of Big Tech with a public and transparent platform for innovation.

With the emergence of energy communities, where a number of prosumers (consumers with their own energy generation) invest in shared renewable generation capacity and battery storage, the issue of fair allocation of benefits and costs has become increasingly important. The Shapley value, a solution concept in cooperative game theory initially proposed by Nobel prize-winning economist Lloyd Shapley, has attracted increasing interest for redistribution in energy settings. However, due to its high time complexity, it is intractable beyond communities of a few dozen prosumers. This study proposes a new deterministic method for approximating the Shapley value in realistic community energy settings and compares its performance with existing methods. To provide a benchmark for the comparisons of these methods, we also design a novel method to compute the exact Shapley value for communities of up to several hundred agents by clustering consumers into a smaller number of demand profiles. Experimental analyses with large-scale case studies of a community of up to 200 household consumers in the UK show that the newly proposed method can achieve very close redistribution to the exact Shapley values but at a much lower (and practically feasible) computation cost. Furthermore, it performed similarly to the probabilistic, state-of-the-art approximation method while having smaller time complexity as well as other desirable characteristics for cost redistribution in energy communities.

The DC Optimal Power Flow (DC-OPF) problem is a widely-studied topic in the field of power systems. A solution to the problem consists of minimizing the running costs of the power system, through defining the optimal operating state for each entity in the system, while adhering to a set of physical constraints. A lot of research has been conducted on decentralized and distributed solutions to this problem, which, when compared to centralized solutions, offer benefits such as adaptability, reliability and scalability. Nevertheless, most of these solutions have only been evaluated through simulations, while physical applications of these algorithms introduce new challenges, such as noise, delays, and the regulation of physical variables like voltage and current. In this thesis, we focus on a decentralized DC-OPF algorithm based on the Consensus and Innovation method, where system entities utilize their physical measurements and communicate with their neighbouring entities in order to reach consensus on a solution. While previous implementations of the algorithm were tested on simulated environments, this thesis explores and proves the effectiveness of the algorithm implemented for a real DC unipolar microgrid, consisting of power supplies, loads and a Power Circuit Board attached to each, where each device's behavior is governed by its own, exclusive entity. The newly-introduced challenges of a physical environment are accounted for, and any negative effects of them are mitigated as much as possible. Furthermore, the algorithm is successfully modified and extended to handle region DC-OPF, something that has not been attempted before, where a single entity of the algorithm could be responsible for many devices in the network. Finally, it is known that the original algorithm has not been experimented on before on scenarios of islanding and de-islanding, which are of importance in OPF, because a power fault may occur and one area may wish to isolate itself from the faulty area, or perhaps a distributed energy resource should only be connected to the network during certain periods, e.g., during sunlight for solar panels. Hence, this thesis also proves the effectivess of the algorithm on scenarios of islanding and de-islanding, in an innovation site called the Green Village, where technologies in the field of sustainable energy provision are tested and applied in a real-life environment.

Computational efficiency is essential for large-scale mathematical optimisation problems, such as the generation expansion planning problem, to be practically applicable. In linear programming solvers, crossover is frequently a bottleneck when solving optimisation problems. This study aims at determining why crossover is a bottleneck in solving the generation expansion planning problem and deriving a method which obtains an optimal solution faster. Symmetry was found to be disrupting crossover, causing its runtime to significantly increase. However, the preceding barrier method benefit from the symmetry handling of the solver. An alternative approach using problem-specific knowledge to reduce symmetry before serving the model to a solver was proposed. The key to the efficiency of this method is using the information of how symmetries are reduced when retrieving the solution to the original problem. The approach was applied to two types of symmetry we identified in the generation expansion planning problem, leading to a significant speed-up of the crossover phase and total runtime for the algorithm resolving a formulation symmetry, but a longer runtime for the method resolving a more practically available non-formulation symmetry. However, the latter was faster in a couple of cases and has many directions for future improvements. The results of this research, thus, demonstrate that domain knowledge of linear programming problems can be applied to reduce symmetry that state-of-the-art solvers cannot detect and to more efficiently obtain an optimal solution than traditional crossover.

Wind energy, generated by windfarms, is playing an increasingly critical role in meeting current and future energy demands. windfarms, however, face a challenge due to the inherent flaw of wake-induced power losses when turbines are located in close proximity. Wakes, characterized by regions of turbulence and lower wind speed, are created as air passes through the rotors, reducing the efficiency of downstream turbines. Wake losses can be reduced by yawing upstream turbines to steer the wake away from downstream turbines. While there are power losses associated with turning turbines off-wind, the gains in the subsequent turbines can outweigh these losses. Yawing turbines to increase overall power output is known as Active Wake Control, and the literature shows that single-agent reinforcement learning algorithms can be used to learn such control policies. However, these approaches are limited to a small number of turbines and do not scale well to larger windfarms. Multi agent reinforcement learning algorithms do scale to larger windfarms and this paper investigates the application of the DGN algorithm for windfarm active wake control. DGN is a fully cooperative algorithm that utilizes a graph representation of agents to encourage collaboration among neighboring turbines. The use of DGN is particularly interesting because windfarms naturally have a topological structure, and depending on the modeling choices, graphs can capture a significant amount of information. This paper demonstrates that DGN can learn useful policies for wake control. Although it does not outperform the DQN single-agent algorithm on small windfarms, its advantage becomes apparent in larger windfarms, where its performance remains consistent while DQN’s performance deteriorates.

Decision-tree evaluation is a widely-used classification approach known for its simplicity and effectiveness. Decision-tree models are shown to be helpful in classifying instances of fraud, malware, or diseases and can be used to make dynamic, flexible access decisions within an access-control system. These applications often require sensitive data of more than one party, like financial or health-related records. It is important to keep this data private, especially when the decision-tree evaluation is done in a collaborative manner where more than one party provides sensitive input data. Current privacy-preserving solutions only consider scenarios in which input data originates from a single source. However, collaboration for decision-tree evaluation tasks is needed more and more since these collaborations often bear fruit. Therefore, in this work we address Private Decision-Tree Evaluation in a collaborative setting. We assume one entity, called the server, that holds the decision tree and multiple users that provide private data on which the decision tree is evaluated. The focus of our research lies on solutions that make use of homomorphic encryption. We give ten different protocols that each take place in a different setting; either the holder of the decision tree receives the evaluation result or the users that provide the input. The protocols use Multi-Key Fully Homomorphic Encryption (FHE) or normal FHE with a Semi-Trusted Third Party (STTP). Additionally, we introduce a novel key-switching method within two of the STTP protocols such that the dependency on the STTP is greatly reduced. All protocols are proven to be secure in the semi-honest model and compared in terms of run-time complexity and communication costs. Due to the high computational overhead for the Multi-Key FHE schemes, the protocols that make use of these schemes are not yet feasible. Therefore, the protocols that use an STTP are the most promising. All protocols take no more than 4 communication rounds. Assuming that the implementation can be parallelized and given an input bit length of 4, the decision-tree evaluation in our protocols takes in the worst case between 60 and 160 hours, executed on an Intel Core Processor at 2.4 GHz with 16 GB memory.

Deep convolutional neural networks (CNNs) have achieved current state-of-the-art in image denoising, but require large datasets for training. Their performance remains limited on smaller real-noise datasets. In this paper, we investigate robust deep learning denoising using transfer learning. We explore the impact of dataset sizes, CNN parameter updates, and noise distribution similarity. Our findings demonstrate that finetuning the decoder while fixing the encoder of an encoder-decoder network architecture avoids overfitting during transfer learning. Moreover, we introduce the concept of noise similarity in transfer learning, showing that reducing the similarity distance between pre-training and finetuning noise significantly enhances CNN performance in denoising. Our results are demonstrated on multiple datasets and various noise camera models. Finally, we validate the robustness and applicability of our approach on real-noise images. All our results and analyses hold and generalize across different datasets, highlighting our insights into the potential of transfer learning for image denoising.

Increased use of the distribution grid due to the uptake of distributed energy resources and the expected penetrations of Electrical Vehicles (EVs) could lead to congestion problems in the distribution grid. Congestion refers to issues related to the overheating of components or voltage issues in the distribution network. Avoiding these issues is crucial to maintain a stable, economical and reliable electricity grid. By using the flexibility of aggregated EVs large investments in grid reinforcement can be avoided. However, a holistic approach is necessary to manage the procurement of flexibility services for all stakeholders involved. One approach is the Universal Smart Energy Framework (USEF), a framework that integrates the existing electricity market with a market for flexibility services from the aggregator to the Distribution System Operator (DSO). This master thesis presents a study on the flexibility market as described by USEF from the perspective of a commercial aggregator of EVs. USEF presents a framework in which flexibility potentially provides financial opportunities for aggregators of EVs. However, it is not clear what the financial impact on the charging costs of an aggregator of EVs is and in what way an aggregator has to adapt its charging logic when trading with DSOs. Therefore, the aim of this thesis is to answer the following question: how can an aggregator of EVs offer flexibility services to a USEF compliant market at distribution level? This report presents an in-depth analysis on USEF, determines the impact of network constraints from USEF on the charging costs of an aggregator of EVs and improves the charging strategy under USEF constraints.

When multiple wind turbines are positioned close to one another, such as in a wind farm, wind turbines located downwind of other turbines are not 100% efficient due to wakes, negatively affecting the total power output of the wind farm. A way to mitigate the loss of power is to steer the wake away from the next turbine, which lowers the current turbine's power output but increases the turbine's total power output. As the number of wind turbines increases, how complex it is to calculate the optimal steering increases exponentially. Reinforcement learning techniques have been a promising candidate to solve this problem. However, single-agent techniques are still very computationally expensive when the number of turbines is the same as an average wind farm. Therefore, this paper aims to see how the QMIX algorithm, a multi-agent reinforcement learning technique, can be efficiently applied to the problem of active wake control. QMIX will be compared to the FLORIS model and a single agent deep reinforcement learning technique TD3 to see if it achieves a higher average reward and converges faster. Finally, QMIX is tested on a larger wind farm to see if it achieves any results in a reasonable amount of time, showing that multi-agent reinforcement learning techniques are more suitable to the problem. This paper shows QMIX has the potential to outperform TD3; although FLORIS is better for smaller wind farms but more research has to be done in applying QMIX to the problem.

Electrical load forecasting, namely short-term load forecasting, is essential to power grids’ safe and efficient operations. The need for accurate short-term load forecasting becomes increasingly pressing with increased renewable energy sources, which are stochastic in their power supply. Most forecasting models are focused on the temporal information for predictions, ignoring the spatial information of neighbouring houses. However, as neighbouring houses are often under similar circumstances, such as weather and holiday conditions, predictions could benefit from this information. Moreover, aggregating electric loads could further improve the accuracy of predictions, as the loads become less stochastic when aggregated. This paper looks at Graph WaveNet, which can capture the hidden spatial dependencies of different houses without needing geographic information about the houses. The framework is compared against a baseline on different aggregation levels. The results show that the framework can benefit from aggregating the residential electrical loads and improves over the baseline on all aggregation levels.

The close proximity of wind turbines to one another in a wind farm can lead to inefficiency in terms of power production due to wake effects. One technique to mitigate the losses is to veer from their individual optimal direction. As such, the wakes can be steered away from downstream turbines in order to increase the overall power output. Multi-Agent Reinforcement Learning (MARL) models the interactions between wind turbines and determines an optimal control strategy through agents that learn the collective consequences of their actions. To analyse the benefit of multi-agent cooperation and centralised critic evaluation, I investigated the effect of Counterfactual Multi-Agent Policy Gradients (COMA) on Active Wake Control. Ultimately, experiments on wind farms of three and sixteen turbines indicate that the algorithm performs moderately, yet worse than single-agent Reinforcement Learning. In addition, high computation costs hinder its application on real-life environments.

The ever-evolving power grid is becoming smarter and smarter. Modern houses come with smart meters and energy conscious consumers will buy additional smart meters to place in their home to help monitor their energy consumption. This new smart technology also opens the door to more accurate power consumption forecasting. In this study we look at utilizing a partial hierarchy, in which one of the appliances in a household is modelled separately from the rest of the house, to help improve household energy consumption forecasting accuracy. This is done in conjunction with Auto Regressive Moving Average (ARMA) based models. Three variants of ARMA based models will be looked at: Auto Regressive Integrated Moving Average (ARIMA), Seasonal Auto Regressive Integrated Moving Average (SARIMA), and Auto Regressive Integrated Moving Average with Exogenous variables (ARIMAX). These methods will then be compared to more baseline approaches such as a persistence method and a seasonal moving average. Our analysis has led us to conclude that the partial hierarchy model offers little to no benefit when applied in the field of household energy consumption forecasting when built upon ARMA based models. ARMA based models in general appeared to be poor performers when it came to household energy consumption forecasting.

This M.Sc. thesis report investigates the application of one-class classification techniques to complex high-dimensional data. The aim of a one-class classifier is to separate target data from non-target data, but only a dataset containing target data is available for training. The issue with high-dimensional data is that it is difficult to perform density estimation due to the `curse of dimensionality'. Most conventional method for one-class classification rely on density estimation. This thesis focusses on the use of autoencoders and generative adversarial networks (GANs) for one-class classification problems involving image data. Autoencoders can learn encoding and decoding functions for samples from the target dataset. These encoding and decoding functions are, however, expected to not perform well for non-target samples, as they have never been seen during the training phase. This makes it possible to separate target and non-target data. For GANs, the discriminator is used to distinguish between target and non-target data. Autoencoders and GANs are evaluated extensively in this report. Their behavior, desired parameters and strengths and weaknesses are evaluated by performing experiments. The main findings are that GANs do not perform well for one-class classification tasks, because of mode collapse and insufficient sampling of the non-target data. Even for extremely simple datasets these issues were observed. Autoencoders are shown to perform much better and behave according to the theoretical expectations.

Non-Intrusive Load Monitoring (NILM) is a technique used to disaggregate household power consumption data into individual appliance components without the need for dedicated meters for each appliance. This paper focuses on improving the generalizability of NILM algorithms to unseen households using Convolutional Neural Networks (CNNs) and one-shot transfer learning. The research investigates the effectiveness of one-shot transfer learning in fine-tuning a CNN model to accurately detect the ON/OFF state of appliances in households not seen during the training phase of the CNN. The study utilizes the Pecan Street dataset for training and evaluation, which includes detailed energy consumption records from various locations in the United States. The results suggest that one-shot transfer learning could enhance the performance of the NILM algorithm, particularly when multiple data samples are used for fine-tuning. However, the effectiveness of one-shot transfer learning varies strongly depending on the number of samples and the characteristics of the target household.

Non-intrusive load monitoring (NILM) is a well-researched concept that aims to provide insights into individual appliance energy usage without the need for dedicated meters. This paper explores the possibility of applying the NILM concept to disaggregate energy data from a community level to a household level. By doing so, it addresses privacy concerns associated with real-time household data collection through smart meters. The proposed approach leverages sequence-to-point learning, a deep learning method, to perform the disaggregation. A method that showed promising results in the domain of NILM. The results show good performance on the evaluation metrics, but they also show that the model does a poor job of capturing the load signature.