http://hdl.handle.net/2268.2/16591 Sun, 05 Apr 2026 19:04:26 GMT 2026-04-05T19:04:26Z http://hdl.handle.net/2268.2/24784 Title: Imbalance Price Forecasting in Belgium under the European Balancing Platform Abstract: This research work describes the development and application of several forecasting techniques to try to predict the belgian imbalance price. The imbalance price is a price that gets applied to certain large actors of the power grid when they are not consuming or producing the expected amount of power. Since this price gets settled ex post when the overall grid imbalance is known, grid actors can benefit from predicting it. This way they can make adjustements before the price gets settled. This is done using machine learning and other forecasting techniques. By design, ELIA, the Belgian transmission system operator wants these actors to adjust their positions to help balance the grid. The problem is that the European Union imposed to its members to join their common balancing platforms. These platforms allow countries to work together when it comes to balancing but this adds a layer of complexity for balancing and also forecasting the imbalance price. The objectives of this work are to see how the regulation changes affect the forecasts and to develop an algorithm that works with these new platforms. To do this, thorough research on the new platforms was done to understand how they work, what they bring and what they will impact. The literature of imbalance price forecasting and surrounding domains was also reviewed. Then, with the help of Flexide’s expertise, the development of forecasting methods began by selecting features in the data, comparing algorithm performances and finally selecting the XGBoost tree ensemble method in the end. This method was then further tuned with different approaches. Finally, the results are presented using a battery simulation that charges and discharges depending on the predicted imbalance price. This gives a first idea of the gains achievable by forecasting. The methods developed in this work managed to slightly outperform the benchmark method that was previously used by Flexide. In conclusion, forecasting the imbalance price may seem like it comes down to predicting chaos at first, but in the right context, forecasts have proven to be good enough to be useful and they are a key to improving the grid balance in the future. Sun, 07 Sep 2025 22:00:00 GMT http://hdl.handle.net/2268.2/24784 2025-09-07T22:00:00Z http://hdl.handle.net/2268.2/23274 Title: Heuristic Methods and Machine Learning for Distribution Network Reconfiguration Abstract: The problem of distribution network reconfiguration (DNR) has been addressed for some fourty years in scientific literature and industry. The aim of reconfiguration is to find the best radial operational configuration in a given electrical state. The main objectives are to reduce active losses and improve the profile. The reconfiguration problem is combinatorial and non-linear, making it practically impossible to solve for real-size electrical networks. To counter this problem, numerous heuristics and metaheuristics have been proposed in the literature. In this thesis, we explore and compare different strategies of reconfiguration. The first method used is based on a minimum spanning tree algorithm to generate an initial solution. This solution is then refined using two heuristics: local search (LS) and tabu search (TS). Finally, a machine learning model is trained to mimic the behavior of these heuristics. This model uses various features to predict whether or not a line is part of the optimal configuration. All methods are tested on different scenarios with and without distributed generators from the IEEE 33 and IEEE 69 networks. Results show that combining classical heuristics with learning-based approaches provides a balance between performance and computational cost, especially for applications requiring fast decision-making. This work contributes to ongoing research into the reconfiguration of constantly evolving electrical distribution networks. Sun, 29 Jun 2025 22:00:00 GMT http://hdl.handle.net/2268.2/23274 2025-06-29T22:00:00Z http://hdl.handle.net/2268.2/20436 Title: Securing Energy Futures: Exploring Belgium's Capacity Remuneration Mechanism Abstract: This thesis explores the integration of low-voltage assets into Belgium’s Capacity Remuneration Mechanism (CRM), focusing on ThermoVault, a company that retrofits residential thermal appliances with intelligent IoT modules. The primary objective is to assess the feasibility and benefits of ThermoVault’s participation in the Y-4 auction for the 2028-2029 delivery period. The research includes an assessment about the need for capacity mechanisms and an overview of European capacity remuneration mechanisms, providing context for the CRM’s design. The Belgian CRM is then explored more in detail, leading to an analysis of ThermoVault’s possible participation in it. The analysis utilizes historical data from a pool of electric storage water heaters (ESWH) over 2023 to evaluate different baseline methodologies, focusing on the "High X of Y " method currently used in Belgium. Results indicate that ThermoVault’s assets can meet CRM requirements, although the unit remuneration values are relatively low. This necessitates an internal evaluation of whether the administrative efforts are justified. The study highlights the need to explore alternative baseline methodologies, such as the declarative baseline methodology, and suggests investigating the remuneration potential for other low-voltage Capacity Market Units (CMUs) in future research. This thesis offers insights into integrating low-voltage assets into capacity markets, providing valuable information for policymakers, industry stakeholders, and companies like ThermoVault in enhancing grid reliability and sustainability. Sun, 23 Jun 2024 22:00:00 GMT http://hdl.handle.net/2268.2/20436 2024-06-23T22:00:00Z http://hdl.handle.net/2268.2/18343 Title: Modeling and Design of a Rural Electrification Model with Microgrids using the GBOML: Case of Kinshasa Abstract: Dans de nombreux pays sous-développés, un problème courant découle du manque d'accès à l'électricité. Ce mémoire de fin d’études aborde la situation électrique en République démocratique du Congo , un pays au cœur de l'Afrique. L'accent est mis ici sur la création d'un modèle visant à fournir de l'électricité aux zones rurales en tenant compte des besoins énergétiques actuels du pays et de la croissance démographique. Pour atteindre cet objectif, nous évaluons le système de distribution d'électricité du pays, explorons son potentiel énergétique et discutons des techniques potentielles d'électrification rurale. De plus, une analyse de la demande en électricité est réalisé dans la localité municipalité pilote de Maluku à Kinshasa. Cependant, en raison de limitations de données, des prédictions précises de consommation n'ont pas été possibles. À la place, nous avons développé et mis en œuvre différents scénarios au sein d'un algorithme d'optimisation. Les résultats obtenus à partir de ces scénarios étaient remarquablement prometteurs. Sur la base de ces constatations, nous pouvons dire que les micro-réseaux connectés aux réseaux principaux pourraient offrir une solution pratique, abordable et durable pour répondre à la demande en électricité dans le pays. Sun, 03 Sep 2023 22:00:00 GMT http://hdl.handle.net/2268.2/18343 2023-09-03T22:00:00Z