< Back to previous page

Project

Resource Adequacy for Modern Power Systems with a Focus on Energy-limited Resources

This dissertation investigates (1) how energy-limited resources, such as battery storage, affect the use of cost-minimising operational and planning models for resource adequacy and (2) how optimisation-based capacity expansion planning models can be improved to accommodate high penetrations of variable renewable energy sources such as wind and solar. The context motivating this research is the increasing volume of variable renewable generation in the European electric power system and the growing adoption of technologies, such as battery and chemical storage solutions, to complement this generation.

Resource adequacy is the ability of an electric power system to supply consumers' aggregate electric power and energy requirements at all times, taking into account scheduled and unscheduled outages of system components. Unlike thermal generators such as gas power plants, there may be many ways of operating energy-limited resources that result in different adequacy indicators \emph{while still being cost optimal}. This dissertation quantifies this effect, which is of the same order of magnitude as the uncertainty inherent in adequacy assessments. It shows that the analytical expression for the optimal level of adequacy, which EU member states are legally required to use, prescribes a peak-shaving operation of energy-limited resources. This operation must be consistently applied across all elements of the resource adequacy framework, namely the reliability standard calculation, adequacy assessments, and capacity mechanisms. This consistency is crucial for the framework to maintain coherence and function as intended. Inconsistent storage operation could result in an economically inefficient power system with insufficient or excessive capacity.

Capacity expansion planning models are frequently used to aid decision-makers in shaping the transition towards a low-carbon energy or power system. These models typically make assumptions whose validity is questionable for future power systems. For example, they will use only a few representative days of data and neglect intertemporal constraints between these days. Doing so fails to capture the value of long-term storage for balancing highly renewable systems. This dissertation compares novel time series aggregation methods and model formulations that use representative days but are still able to value long-term storage. While solutions exist, they may still require more representative days than when long-term storage is absent. Greater volumes of variable renewable generation will require more operating reserves to deal with forecast errors. This dissertation proposes a novel, computationally-efficient method of accounting for this in capacity expansion planning models.

Date:11 Mar 2019 →  8 Mar 2024
Keywords:operating reserves, resource adequacy, energy-limited resources, capacity expansion planning, storage, reliability, security of supply
Disciplines:Electrical energy production and distribution, Renewable power and energy systems engineering
Project type:PhD project