Flexibility and uncertainty in power systems:Insights from Electric Vehicle Charging

The increasing share of renewable generation, rapid electrification of energy services, and growing adoption of distributed energy resources make the energy system more dynamic and uncertain, fundamentally complicating the continuous balancing of supply and demand required to maintain stability and reliability. As the energy transition advances, the variability and uncertainty in both supply and demand have reached levels such that conventional solutions, such as grid reinforcement, are no longer sufficient to prevent congestion. Consequently, there is a growing focus on demand-side flexibility as a complementary means of addressing these challenges. Among the various sources of demand-side flexibility, electric vehicles (EVs) have emerged as a particularly promising option. Their flexibility arises because EVs are typically parked for much longer periods than required for charging, enabling operators to shift charging over time through smart charging. While EVs are widely recognized for their potential to provide flexibility, much less is understood about the conditions under which this potential can be effectively operationalized in the presence of uncertainty. Developing this understanding is crucial, as operators and aggregators must decide how, when, and where to deploy flexible resources to balance supply and demand or hedge against market price risks. This dissertation examines different approaches to modeling demand-side flexibility, specifically in the form of smart charging, through the lens of inventory management. It aims to strike a balance between realism and tractability, enabling a more precise understanding of how flexibility operates under uncertainty.