Decision-making under uncertainty in short-term electricity markets

In the course of the energy transition, the share of electricity generation from renewable energy sources in Germany has increased significantly in recent years and will continue to rise. Particularly fluctuating renewables like wind and solar bring more uncertainty and volatility to the electricity system. As markets determine the unit commitment in systems with self-dispatch, many changes have been made to the design of electricity markets to meet the new challenges. Thereby, a trend towards real-time can be observed. Short-term electricity markets are becoming more important and are seen as suitable for efficient resource allocation. Therefore, it is inevitable for market participants to develop strategies for trading electricity and flexibility in these segments.

The research conducted in this thesis aims to enable better decisions in short-term electricity markets. To achieve this, a multitude of quantitative methods is developed and applied: (a) forecasting methods based on econometrics and machine learning, (b) methods for stochastic modeling of time series, (c) scenario generation and reduction methods, as well as (d) stochastic programming methods. ... mehrMost significantly, two- and three-stage stochastic optimization problems are formulated to derive optimal trading decisions and unit commitment in the context of short-term electricity markets. The problem formulations adequately account for the sequential structure, the characteristics and the technical requirements of the different market segments, as well as the available information regarding uncertain generation volumes and prices. The thesis contains three case studies focusing on the German electricity markets.

Results confirm that, based on appropriate representations of the uncertainty of market prices and renewable generation, the optimization approaches allow to derive sound trading strategies across multiple revenue streams, with which market participants can effectively balance the inevitable trade-off between expected profit and associated risk. By considering coherent risk metrics and flexibly adaptable risk attitudes, the trading strategies allow to substantially reduce risk with only moderate expected profit losses. These results are significant, as improving trading decisions that determine the allocation of resources in the electricity system plays a key role in coping with the uncertainty from renewables and hence contributes to the ultimate success of the energy transition.