Local Energy Trading for Microgrids - Modeling Human Behavior, Uncertainty and Grid Constraints
Jens Hönen is a PhD student in the department Mathematics of Operations Research. Promotors are prof.dr. J.L. Hurink from the faculty of Electrical Engineering, Mathematics and Computer Science and prof.dr. A.P. Zwart from Eindhoven University of Technology.
Electricity is one of the major drivers of today’s society and a crucial element of the rapid development of humankind over the last 150 years. However, it also significantly contributes to the ongoing global warming, a potential threat to the current and future society. To reduce its impact, the energy transition aims to change the electricity production by switching from fossil fuels to more renewable and sustainable energy sources. In addition, the energy transition also addresses large changes in the heating and mobility sector, where fossil fuels are replaced by electricity.
These changes have a significant impact on the electricity system, and an intelligent (active) management of the electricity production and consumption is required. Throughout the last few years, many approaches for such active management have been proposed.
At the core of this thesis, we focus on three different aspects of such approaches
1: We consider and analyze the impact of human behavior on the outcome of a local electricity market. Due to direct participation of households, the question arises of how human preferences and behavior affect the outcome of a local electricity market. Therefore, we translate a behavioral model from social science into a multi-objective optimization problem, which uses the personal preferences and motives of households to create tailor-made bidcurves. We analyze the results on a household and the market level to derive implications for future market design.
2: The second aspect of this thesis concerns the increased uncertainties in the future energy system due to the intermittent nature of renewable energy sources. To deal with such uncertainties, we focus on a joint energy management of a neighborhood. We apply ideas and techniques from robust optimization to deal with the uncertainty and mainly focus on an approach combining static robust optimization with a rolling horizon framework. Hereby, we first analyze the impact of different sources of uncertainty on the objective value. Based on the gained insights, we generalize the rolling horizon framework by allowing more flexible starting time slots and compare and analyze two such generalized rolling horizon versions.
3: The third aspect focuses on the impact of energy trading on the electricity grid. The increased peaks due to the additional production and generation pose a serious burden to the current electricity grid. To ensure a safe operation, we focus on grid constraints in the context of a real-time control approach, which implements day-ahead and intraday market solutions. We use the planned solutions to guide the real-time decisions and propose a three-step online algorithm. In addition, we identify an interesting connection between day-ahead operations and their real-time realization.
