An Adequate Level of Resilience

Each year, and across the U.S. power grid, unexpected, extended electricity power disruptions cause extensive physical harm and inconvenience to communities and individuals, and billions of dollars in direct and indirect economic losses, consequences of arguably inadequate levels of electrical system resiliency.

From a distribution grid investment planning point of view, there is a yawning gap between our general awareness of the potential dangers posed by the next big storm (or other causes) and our ability to economically value its consequences. Because it's hard, and the tools we do have are imprecise, it might be getting conveniently pushed to the side. But is this prudent? 

Improving resiliency certainly isn't just a matter for the distribution grid, (concerning as it does the bulk power system and the reality of impacts to generation and transmission assets). The resiliency valuation challenges are lurking within most grid plans if we look closely enough.

There's jointness: whereas most grid plans promote improvements in order to improve reliability, such as feeder hardening, more sophisticated fuses and reclosers, and advanced communications and software, these same improvements contain a fair measure of resiliency impacting improvements at the same time.

On the Table, Regardless

Resiliency is on the table whether we call it out or not. It should spark the same questions we throw at reliability when addressing cost-of-service and prudent investments. What about priorities, tradeoffs, impact estimates, costs, benefits, and risks, inclusive of resiliency considerations? We can't rigorously construct a distribution grid plan without answering: Why this, why not that?

When wide-scale system interruptions occur, the loss or erosion of resiliency is revealed in both the physical and economic damages that ensue. These typically greatly exceed those experienced as part of normal, day-to-day, reliability scaled events. One major east coast utility storm event caused fifteen to twenty times the level of outage minutes compared to the sum of all reliability scaled events across a five-year period.

Despite this disproportionate level of impact, grid plans are often sponsored showing hundreds of millions of dollar value of reliability improvements without mention of resiliency improvement value. If the same grid reinforcements improve resiliency, there is easily several times as much economic value on the table, but the plans mainly give a qualitative shout out to resiliency's commensurate level of physical and economic impacts.

The Grid Will Change and Expand

The resiliency valuation challenge matters a great deal. The grid is aging. We are spending sizable and always-scarce resources on improving it. The risk envelope is expanding, with more disruptive weather, the role of bad actors (such as cyber threats, terrorism), and the growing complexity of grid operations.

The recent flooding of several eastern counties of Kentucky point to climate's volatility bringing more damaging events, more grid damages, and catastrophic levels of disruption to communities and livelihoods.

Andrew Trump: Valuing resiliency is important, if not essential, for determining the best directions of our future grid investments.

There is also a huge push to electrify more of our lives, putting unrelenting pressure to expand the grid's capacity at every level. This includes new demands, such as EVs, and means that people expect ever greater levels of electrical capacity, convenience, and reliability. Just one residential EV can easily add four thousand kilowatt-hours of energy supply requirements to a typical residence, on top of the eleven thousand kilowatt-hours already needed.

Recent studies in California, Massachusetts, and the D.C. region point to a huge expansion of grid capacity if we want to electrify more and more energy services.

Bookends: Adaptation

In the face of these grid expansion challenges, how can problems be approached differently? How do we adapt versus just rebuild? Should a shoreline community rebuild with stronger poles, or should zoning changes force development in new directions altogether? Should we build to a thousand-year standard? 

Adaptation thinking is essential for us to make good choices, but we won't be able to do it well unless we develop greater facility, confidence, and practice discipline in figuring out the physical and economic damages, as well as mitigations related to resiliency. How can we sort out the choices between restoring versus adapting if we don't understand the risks, costs, and benefits of addressing resiliency?

Language Matters

Caleb Kao: How can we sort out the choices between restoring versus adapting if we don’t understand the risks, costs, and benefits of addressing resiliency?

Resiliency has strong emotional appeal as a concept; many of us are very comfortable in our understanding of it when applied to our non-professional walks of life. We are drawn to the emotional appeal of resiliency when communities face the aftermath of a tragedy, or when witnessing our kids standing up to life's big challenges. It's about withstanding difficulties and when something bad happens, springing back and getting back on our feet.

It's the same with our reactions to the apparent need for a more resilient grid. The psychologist Daniel Kahneman describes how certain words "easily trigger coherent patterns of activated ideas in associated memory" — but resiliency also comes along with strong biases. It may crowd out ambiguity. Grid resiliency is such an obvious need and grid attribute, in other words, why bother with exploring its details through disciplined and structured analysis?

Toward a Working Definition

This is lazy thinking. As soon as we start to peel the resiliency onion back, we run into a host of technical evaluation challenges. A quick read of a well put together IEEE technical subcommittee report on resiliency starts to reveal these.

A lot emerges out of what we already know and practice related to reliability.

Resiliency is a special condition of reliability, and it's doubtful we would want to jettison a huge body of energy law, policy, and procedure that guides and directs reliability for the sake of defining resiliency.

Take, for example, FERC's important Order 2222 (Final Rule, September 17, 2020) on the market role of distributed energy resources. It mentions resiliency five times within its two hundred ninety pages. It is also always paired with reliability, as in reliability and resiliency or resilience and reliability. Resiliency is never used separately.

This makes sense because we are dealing with outage conditions. We want to borrow as much as we can from a well-established practice discipline for reliability. But resiliency's special role is that it aims to deal with high impact, infrequent events — the conditions that are often the hardest to address as part of risk assessment.

Assessing risk always involves addressing these pairs — what's the nature of the impacts? How often might they occur? Estimating these two risk attributes is a big part of what makes valuing resiliency difficult. But that's the nature of risk assessment, and it is never easy.

While this sets up some basics, there is a lot more involved in determining an approach for valuing resiliency as a special reliability condition. The challenges relate to how to deal with the black box nature of the best reliability valuation tools we currently have.

These tools can be useful. They provide insights and context, but are not perfect solutions by any means, in part because the underlying algorithms used in the black box are not intended to reach into long duration events beyond a few hours, or to cover the full gamut of direct and indirect consequences of big, impactful multi-day system disruptions.

The challenges also relate to the nature of cost-of-service and prudency reviews, which are typically a form of extended dialogue mainly originating out of a straightforward, scenario-based cost and benefit formulation. The world gets reduced to two scenarios.

One involves what we know about today. The other addresses a future state. This simple cost-benefit formulation has a hard time ingesting the messy prongs of risk assessment — consequences and likelihoods.

This is an oil and water problem that sets up problems of incompleteness rattling around the confines of our efforts to determine the reasonableness and prudence of utility investments.

Paths Forward

Valuing resiliency is important, if not essential for determining the best directions of our future grid investments. Challenges can be overcome with some flexibility of reason and judgment proportionate to the evaluation challenges at hand. We are at a very early stage in progressing the tools and disciplines that are needed, however.

Things start by admitting to this state of affairs and agreeing to what we don't know.