Investing in Power System Resilience: A mixed methods approach to assessing the tradeoffs of resilience strategies

As modern American society has become increasingly dependent on the provision of reliable electric power, the importance of enhancing the resilience of the power system has grown. In years to come, stress arising from climate change, and the continuing risks of physical and cyberattacks on the system, will likely make resilience even more important. However, while the topic receives a great deal of attention, there is little rigorous analytical work that has assessed the efficacy and cost-effectiveness of strategies to increase power system resilience. 

This research adopts a mixed methods approach to identify and assess strategies for improving the resilience of power systems and thus enhance planning as the power system evolves. Chapter 2 begins by reviewing a range of resilience strategies discussed in the literature that could be deployed on the grid. The literature on this subject is scattered and unsystematic in its treatment of cost and efficacy. This chapter contributes to the resilience literature by developing a comprehensive list of strategies and summarizing what is known about their costs and efficacy. 

Chapter 3 has already been published in the journal Risk Analysis. It analyzes and compares a range of individual and collective strategies that could increase the resilience of power supply to residential customers on a distribution feeder by providing contingency power during large power outages of long duration. A typical (but hypothetical) town in the Upper Connecticut River Valley is modelled—we choose this region because it is susceptible to ice storms and hurricanes—to estimate the cost and performance of different resilience strategies, assuming that a large power outage of long duration occurs. This work showed the extent to which collective strategies are cheaper, and therefore highlighted the importance of developing institutional arrangements that make it easier for communities to implement those collective strategies.

Chapters 4 and 5 present the first systematic analysis of empirical data about the efficacy of different resilience strategies. To do this, Chapter 4 focuses on the state of Florida, which has required utilities to make substantial investments in storm hardening to improve the resilience of their systems in the face of tropical cyclones. While these utilities compile and report to the state Public Service Commission a phenomenal amount of data on the cost and performance of their different resilience strategies, almost no analysis has been done on these data to assess whether and to what extent these investments are actually improving the power system’s resilience. After compiling fifteen years’ worth of these data for Florida’s Investor-Owned Utilities, the exploratory data analysis shows that after accounting for tropical cyclone severity, undergrounding power lines, changing pole material to non-wood and installing advanced metering infrastructure across all customers likely improve average tropical cyclone restoration time (CAIDI). Increasing annual tree trimming on the other hand does not seem to have an impact on tropical cyclone CAIDI. However, to continue performing at current levels of reliability, these utilities will have to spend more on the same activities and even more on others to continue to address the impacts of climate change and the uncertainty they induce. 

Informed in part by this applied data analysis, in work reported in Chapter 5 utility engineers were interviewed to extract their mental models of power system resilience, as well as their judgments regarding how to enhance resilience most effectively given the range of threats facing their systems. Operators were asked to consider the worst outage their system had experienced and then to elaborate on how implementing resilience strategies could have improved their response to this worst outage without financial barriers. Undergrounding power lines was the preferred strategy by interviewees, but also estimated to be the most expensive. Finally, participants were asked what technologies, policies and resources are especially needed by utilities to enhance their systems’ resilience, and they suggested new rate designs to account for an increase in renewables, distributed energy resources, and electrification as well as defining new policy standards for new loads and distributed generation being added to their systems. They were additionally concerned with supply chain backlogs and having enough money to do the necessary upgrades on an aging system.