Unit Commitment Models and Algorithms for the Simulation of the European Power System under many Weather Scenarios

Abstract

Mitigation efforts to avoid dangerous effects of climate change are leading to a shift in electricity production towards low carbon and intermittent renewable energy sources. This shift, along with the increasing use of electric heating and transportation, introduces more variable generation and demand in the power system that are both influenced by weather patterns. To assess the adequacy of future power systems in Europe and identify potential problems, power system models are needed to simulate their operation under many weather scenarios. The unit commitment problem (UC) is a well-known optimization problem that is often used for simulating power system operation. Solving these UC problems is challenging since it is a hard optimization problem, and the size of the power system models makes solving it even harder. This results in incredibly high computation times which can be addressed by either increasing the efficiency of the algorithms or performing justified simplifications to the UC problem. This thesis improves power system modeling based on the UC by investigating the impact of modeling decisions, making algorithmic enhancements, and investigating the relationship between power system adequacy and weather regimes of Europe by performing power system modeling based on the UC.