Online

Kelly-Gorham, Molly Rose

Electrical Engineering

2022

Ph. D.

Concerns about climate change are leading to a transformation of energy systems, including increased adoption of renewable and distributed power generation and energy storage. Climate change is also increasing uncertainties in the environmental conditions in which electric power systems operate and increasing regional vulnerability to extreme events. Under this variable and uncertain environment, there is a need for research that identifies high risk vulnerabilities and that identifies technology and policies that most effectively improve resilience. However, quantifying resilience is hard. No single number can fully describe the resilience of a particular system, and there is as of yet no consensus about the most effective methods for estimating the resilience of electric power systems. This dissertation contains three core chapters that (1) analyze energy usage and resilience in Vermont and the policies shaping the state's progress towards its energy goals, (2) create a framework to measure resilience using data driven methods to conserve accuracy, model simply, and broaden the range of events examined, and (3) quantify the effects of interdependence on power system resilience. Chapter 1, the introduction, provides the motivation, a literature review, and a summary of the gaps in the literature on energy/infrastructure resilience, which this dissertation aims to address. Chapter 2 provides a qualitative look at the resilience elements of Vermont's decarbonization goals and several policy action recommendations that address the decarbonization and resilience challenge. Chapter 3 presents a new method to measure resilience starting from historical utility data. Application of the method to the IEEE RTS 96 test case utilizing Vermont weather data suggests that distributed solar and storage both improve resilience, and storage is a particularly valuable asset. Chapter 4 improves the method of measuring resilience from Chapter 3 with a new approach to stratified sampling of transmission line and generator outage data to measure low probability, high impact events and a new method of modeling and ranking the impact of interdependencies on power system resilience. The findings from two case studies on a small and a large power system in northeastern US confirm that natural gas supply interactions can have a severe impact on power system resilience. Finally, Chapter 5 concludes the work presented in the previous chapters. In sum, this dissertation improves the current understanding of energy system resilience in the face of climate change.