Titre de la dissertation: « Noncooperative Game Theory for Resources Scheduling and Planning in Renewable Energy Communities ». 

Promoteur de thèse: Monsieur Thomas Brihaye et Co-promoteur Monsieur Zacharie De Grève

Résumé de la dissertation: The events of the last few years, such as the COVID-19 pandemic and the geopolitical crisis which started in Eastern Europe in early 2022, have thrown the natural gas and electricity European markets into an unprecedented crisis. These circumstances have highlighted the need for structural and regulatory measures to protect end-users from market fluctuations, while accelerating the transition to more resilient and sustainable systems. At the same time, the electricity sector is undergoing a profound transformation, with the rise of distribution energy resources and the growing adoption of decentralized solutions such as local solar and wind generation or individual storage systems. The developments reflect a paradigm shift towards more participatory and smarter, consumer-centric energy models. In this context, renewable energy communities are emerging as key actors in the energy transition, and have received particular interest from economic, political and academic sectors in recent years. They are organized entities, gathering consumers and prosumers allowed to exchange renewable electricity produced locally without resorting to the traditional wholesale/retail markets. Their purpose is to provide economic, environmental or social benifits to the members and society, rather than to make a financial profit.The main objective of this thesis is to model renewable energy communities and the various challenges surrounding them using noncooperative game theory.  For that purpose, this work is divided into two parts, exploring a specific problem that can be modeled with a specific noncooperative game form.In the first part of the thesis, we study local energy communities composed by end-users connected to the public electricity distribution network and sharing common resources such as the grid and their own local generation. We propose two market designs for the optimal day-ahead scheduling of energy exchanges within these communities. The first one implements a collaborative demand-side management scheme inside a community where members objectives are coupled through grid tariffs, the second allows the valuation of excess generation in the community and on the retail market. Two grid tariff structures are tested, one academic and one which reflects the current Belgian regulations in terms of grid tariffs. Individuals’ bills are obtained through 4 methods of cost allocation. Both designs are formulated as optimization problems first, and as noncooperative strategic games then. In the latter case, the existence and efficiency of the corresponding (generalized) Nash equilibria are studied and solution algorithms are proposed. The models are tested on a use-case made of 55 members and compared with a benchmark situation where members act individually. We compute the global renewable energy community and members’ individual costs, study the inefficiencies of the decentralized models compared to social optima, and calculate technical indices such as self-consumption or peak-to-average ratio. In addition, we investigate the influence of retail electricity prices on the daily operation of the energy community. A sensitivity analysis is performed on the retail electricity prices and we measure the impact on the total community and members individuals’ costs and interest in joining/leaving the community.The second part focuses on the integration of a new member inside an existing renewable energy community. We propose two distinct approaches. In the first structure, we model the case of an external user interested in joining the community, with ou without investment contribution. The second approach examines the situation where the community is the instigator of its own expansion. This allows us to analyze how the flexibility or thoroughness of integration processes can influence the community dynamics and its ability to remain consistent with its objectives. Long-term (investments and tariff adjustments) and short-term decisions (day-ahead resources scheduling) are handled by an extensive-form game considering the uncertainty linked to the evolution of the retail market price.  In particular, we use the results obtained in the first part to model and solve the short-term level. We also include the case where potential candidates and the community present heterogeneous preferences, reflecting varied objectives and priorities, such as minimizing costs or CO2 emissions, maximizing return on investments, etc. In addition, we compare the decision-making processes of candidate users and the community under uncertainty. Our analysis is based on two distinct theoretical frameworks: (1) expected utility theory, which assumes perfect rationality on the agents’ side, and (2) prospect theory, which captures the bounded rationality of individuals and their biases in risk perception. The models developed were tested on three renewable energy communities with distinct energy profiles and a varied list of candidates. Different combinations of preference criteria and parameters of the decision functions are explored in order to analyze the interactions and impacts of agents’ preferences and perceptions. The models developed in this part have a sufficiently general structure to be extended to other types of decisions and problems, as well as to a variety of stakeholder profiles.