The full report, available here, is authored by:
- Alison Silverstein, Alison Silverstein Consulting – (512) 964-0787, alisonsilverstein@mac.com
- Rob Gramlich, Grid Strategies LLC — (202) 821-6943, rgramlich@gridstrategiesllc.com
- Michael Goggin, Grid Strategies LLC — (202) 302-9670, mgoggin@gridstrategiesllc.com
Highlights of the report include:
Customers pay the ultimate price for power outages, whether through their electric bills or their own personal losses and expenditures. Increasing numbers of bad weather events have led many customers to expect that more outages will happen. Therefore we should measure power system resilience from the end user’s perspective – how many outages happen (frequency), the number of customers affected by an outage (scale), and the length of time before interrupted service can be restored (duration). And since long outages do occur, we should also consider customer survivability as an important element of resilience preparations. (p. 3)
History shows that the vast majority of outage events arise at the distribution and transmission levels due to weather events. The Rhodium Group finds that the bulk of outage events are due to routine causes (local storms, vegetation, squirrels, equipment problems), and the Department of Energy reported that 90% of electric power interruptions arise on the distribution system, mostly weather-related. High-impact, low-frequency events such as hurricanes and winter storms cause about half of customer outage-minutes. At the other end of the probability and causal spectrum, Rhodium determined that less than 0.1% of customer outage-minutes were caused by generation shortfalls or fuel supply over the 2012-2016 period. (pp. 3-4)
Most outage events and threats have common consequences — they damage distribution and transmission assets, causing customers to lose electric service. A proactive approach to reliability and resilience would take an all-hazards approach and focus on how to address and mitigate these common consequences, managing risk by taking measures that mitigate against as many threats as possible. (p. 4)
The best way to assess the cost-effectiveness of a reliability or resilience measure, and compare between measures, is to estimate its impact on the probability of outage frequency, magnitude and duration, and upon customer survivability. A constructive resilience analysis process will define resilience goals, articulate system, and resilience metrics, characterize threats and their probabilities and consequences, and evaluate the effectiveness of alternative resilience measures for avoiding or mitigating the threats. Regulators and stakeholders should ask how each remedy (individually and in suites of solutions) might reduce the frequency, magnitude and duration of customer outages relative to the entire scope of customer outages, not just those resulting from generation- or transmission-level causes. This analysis should be both threat-agnostic and jurisdiction-agnostic – many of the best solutions to maintain and enhance resilience lie beyond the limits of the bulk power system and federal jurisdiction. (p. 5)
Many of the measures that offer the highest value for reliability and resilience delivery address the provision, operation and maintenance of distribution and transmission assets, because those are the power system elements that are most frequently damaged by routine events and severe weather. Most of these T&D measures are effective against a wide range of threats and deliver multiple benefits – for instance, an inventory of critical spare equipment can be used to deal with a variety of damages and causes, emergency planning and exercises improve response effectiveness against many types of disasters, and transmission automation or situational awareness can be used to improve system efficiency and resource integration. Similarly, measures that protect customer survivability, such as more energy efficient building shells and distributed generation with smart inverters (to keep providing energy to the host after the surrounding grid is out of service), help customers under many adverse threats and offer multiple benefits (such as customer bill savings and comfort). (p. 6)
Since most outages occur due to problems at the distribution level and long-duration outages are caused primarily by severe weather events, it logically follows that measures that strengthen distribution and hasten recovery would be highly cost-effective. In contrast, measures to make generation more resilient are likely to have little impact on outage frequency, duration or magnitude or on customer survivability. (p. 7)
There is little current basis for finding that generation supply — as a generic issue — is a serious threat to power system resilience. (p. 34)
Generation and fuel supply shortages rarely cause customer outages, and when they do it is almost always due to an extreme weather event or operational failure that may also affect T&D assets. No single unit or type of generation is critical or resilient in itself. Grid operators have always relied on a portfolio of resources performing diverse roles to meet the range of reliability services needed; over the past decade, those portfolios have expanded to include distributed resources such as demand response and distributed generation. Many alternate portfolios of supply- and demand-side resources can provide reliable power delivery. (p. 6)
To ensure that electricity markets operate efficiently and support reliability, reliability services should be defined in functional, technology-neutral terms based on actual system needs, rather than in terms of the characteristics or attributes of resources that historically provided those services.
To assess generation and fuel security threats, it is important to distinguish system reliability or resilience from plant- or technology-specific reliability or resilience. Power systems utilize a portfolio of resources such that the loss of any one unit can be covered by activating others which are held in reserve. Thus no individual unit or technology is critical, and it is not meaningful to assign a level of “reliability” or “resilience” to a generating unit or a type of generating technology. Rather, all power systems perform system-wide analyses to make sure they have enough aggregate energy and reliability services. … No single resource or technology is essential because all of the needed energy and reliability services can be provided by a wide range of technology combinations, including combinations that include no nuclear, no coal, no gas, or no renewable sources. (pp. 33-34)
The combination of a generation fleet and robust transmission system, with customer-side demand response and distributed generation assets, generally offsets the outage risk from losing individual plants or fuel sources. Because the marginal benefit for customers of protecting generation is quite low (particularly when reserve margins are high), generation-related solutions are generally not the most cost-effective means of reducing customer outages on power systems today. There is no evident need to compensate generators or other assets for bulk power system resilience beyond the engineering-based reliability services already being procured. (p. 7)
Electric utilities and customers must deal with the consequences and costs of rules and decisions intended to foster reliability and resilience, including well-intended policies that crowd out or preclude more useful and impactful investments and actions. Every investment has an opportunity cost. There is a great risk that if regulators and stakeholders do not conduct the type of analyses suggested here to inform and coordinate resilience investments, we will end up committing significant amounts of money and effort to improve resilience, yet have little constructive impact on the probabilities or actual levels of future customer outages. (p. 7)
Filed Under: Uncategorized