Skip to main content

Delivering reliable energy under extreme conditions

Mark Dyson on designing a more resilient power grid

From hurricanes in Puerto Rico to wildfires in California to freezing temperatures in Texas, extreme weather events are happening with greater frequency—and driving widespread failures of the local power grids. Mark Dyson, a principal with the Carbon-Free Electricity Practice at Rocky Mountain Institute, studies grid resilience. Recently, he spoke to us about how to design a more resilient grid.

Long-term, large-area grid outages driven by severe weather alone now cause tens of billions of dollars in damage to the US economy each year, and this threat is only one of many that are growing in impact and likelihood.

Mark Dyson, Reimagining Grid Resilience: A Framework for Addressing Catastrophic Threats to the US Electricity Grid in an Era of Transformational Change

Deloitte: Mark, thanks so much for joining us. Not so long ago, the idea of large populations in this country going without heat and electricity for more than a day or two seemed unimaginable. Now it seems to happen almost every season somewhere in the United States. Why is that?

Mark Dyson (Mark): A few factors contribute to this. First, there’s just more extreme weather than there used to be. It's pretty clear that this is consistent with global average temperature rise. Another factor is aging infrastructure. The United States is behind many other rich countries in investment over the past few decades in its transmission and distribution system. Utilities are belatedly prioritising investment in the grid, but it has lagged behind other developed countries for a number of years now. The third thing is just the growing interconnectedness and increasing centrality of electricity to the economy. In an increasingly digital and increasingly electrified economy, we haven't paid as much attention to the vulnerabilities implied by a fragile grid.

Deloitte: That last point is interesting. Are you saying that connectedness makes the consequences more far-reaching when the power goes out?

Mark: Yes, that’s right. You could really see this interconnectedness of energy with every other kind of major infrastructure system in the recent blackouts in Texas. When the gas wells froze in the Permian Basin in Texas, that cut off pressure to the gas system, which cut off the gas to power plants, which then cut off electricity generation for pumping gas through pipelines, so it was this whole vicious feedback cycle.

Deloitte: That gets to the question of grid design. You’ve written about the weaknesses of a linear system, where there’s a simple progression in delivering electricity from the fuel source through generation, transmission, and distribution to the end user. And your point is that, when there’s a failure anywhere in that line, the effects cascade to every point down the line. How did we end up with that kind of system? Did it ever make sense?

Mark: Well, it probably made more sense 50 years ago than it does today. For a long time, there were a lot of economies of scale in most aspects of the electricity supply chain. For many years, up until about the ’70s, we were really good at making plants bigger, more efficient, and therefore more cost-effective per unit of electricity. And so there was this increasing size of coal generators, for example. Those scale benefits reached their limit in the ’70s, but because that dynamic existed for so long during the birth of our power grid, it locked in a system of centralised generation facilities thatssend power through grids with common points of failure. The system arose out of the economies of scale that existed 50 to 60 years ago, and it locked us into infrastructure choices that we still have to deal with today.

Deloitte: So what is the alternative to that centralised system?

Mark: The first step to improving what we have is to become more efficient so we can minimise the demand on the grid in meeting critical loads. We ran some back-of-the-envelope calculations based on data from the grid operator and found that the implied demand for heating energy in Texas from homes that use electricity for heat was on the order of 35 gigawatts at the peak of the outage. That’s greater than the summer peak demand for air conditioning. Why did that happen? It’s because there wasn't a building code in Texas until recently. There was no requirement to make homes efficient, and there was no requirement to put insulation in walls. And so, many homeowners and developers decided just to not invest in that simple, relatively low-cost option. Investments like that up front would have cut that demand down significantly and allowed a lot more people access to power. It’s important to recognise that this problem doesn’t start at the generator. This problem starts at the source at which you’re demanding power.

There are also design improvements. I’m advocating that we look for opportunities to build distributed resources that are closer to the customer and therefore more resilient to some of the failures that we’re seeing with increasing regularity. This wouldn’t replace what we have; it would be a complement to a system that actually works pretty well—most of the time. There’s an opportunity to plan better for the worst days. That means investing closer to the customer, with distributed resources that can help meet critical need. By that, I mean critical in terms of safety or economic activity in the increasingly likely event of a large-scale outage.

Think of what happened in Texas this winter. A critical load for a customer who lost power would have been just a little bit of power—enough to be able to turn on the furnace so that the homeowner could burn gas that was flowing through the pipes into their house. Not enough to back up their whole house, but enough to run their furnace. That’s a critical load in that instance. And it actually doesn't take that much backup power to run that load.

We’re learning that the grid as a whole faces numerous risks that can shut it down quickly, not only from extreme weather, but from terrorist attack or cyberattack. There are inherent risks to centralised, bigger-is-better system that we can’t get around.

Deloitte: Are you talking about a backup generator at the house, or is that something that could be designed in at the grid level?

Mark: One way would be for homeowners to install a solar PV storage system at their home. It could connect to the home main panel in such a way that it would prioritise critical loads within the home during a contingency event like what we saw in Texas. Maybe that would be heat or heat plus the refrigerator, or heat plus one light source. Another way would be a simple backup generator.

So those are two options at the home level that policymakers could help homeowners do, just like they’re doing in California to prepare for power outages during wildfires. There, state policymakers are helping customers invest in solar power storage so they can keep some critical services active during those future outages that we know are coming.

Another really interesting approach that one of the utilities I’ve worked with here in Colorado is pursuing is at the community level, where, by using an intelligent grid control, they can allow their customers to prioritise a small, critical amount of electricity during emergencies—enough to keep everybody relatively warm and safe during a disruption in supply. So, everybody has access to some power, and nobody loses power completely.

Deloitte: Can you talk a little about why small home and community systems are better than a big, professionally managed, centralised system? What is the advantage?

Mark: We’re learning that the grid as a whole faces numerous risks that can shut it down quickly, not only from extreme weather, but from terrorist attack or cyberattack. Solar storms can disable large swaths of the grid without warning. A single strike can shut off power to large areas. There are inherent risks to this centralised, bigger-is-better system that we can’t get around. The system as, designed today, is brittle.

The smaller-scale, closer-to-consumers version has some advantages. One is that hardening the main grid so that it is more resistant to outages just costs money, and it never, ever provides economic value. It’s like buying insurance. You pay a premium every year, and you don’t get anything back unless the unthinkable happens. That's the same principle here, that we should continue buying insurance and we should continue hardening the grid against possible disruptions.

But the options that are closer to the customer often actually provide economic value on a day-to-day basis, even when the main grid is fully functional. For example, being more energy-efficient saves money every day of the year for consumers and for the grid. Solar plus storage generates power every day that the sun is shining, and the storage component can help reduce overall system demand and therefore limit the need to invest in additional power plants to meet peak demand. These relatively small-scale, close-to-the-customer interventions provide resilience in outage conditions and economic value every day of the year.

Deloitte: Why have utilities and power companies been slow to adopt any of these principles?

Mark: Because, in many cases, these principles imply an existential threat to their business model and their shareholders. The more customers buy solar panels and put them on their roofs and buy batteries, the lower their reliance on the grid—and the lower the revenue that the power company collects from each customer. And, of course, this is a big topic, and there’s a lot of nuance here. But at the end of the day, it’s not good for the power company to have their customers lower their reliance. And that, I think, is the fundamental reason that these operations have been sidelined.

Deloitte: The business school response to that would be that power companies should immediately get into the home solar business so they can disrupt their own businesses rather than letting someone else do it.

Mark: They’re trying. We’ve seen some action there, but they’re limited in their ability to do that because the solar companies don’t want the incumbent monopoly to compete against them. The utilities have an advantage, given their 100 years of being in business and the customer relationships they’ve identified. And with their monopoly position, it’s not necessarily good for competition, even if it's good for the utility’s business model.

Deloitte: Mark, this has been a fantastic discussion. Is there anything else you’d like to add?

Mark: I'll just conclude with the thought that in the 2020s, however you slice it, there’s going to be about a trillion dollars of spending, of capital investment in the US power system. So, we’re at an inflection point. We’ve seen these horrific outages that have reinforced the vulnerabilities of the current system. From both a resilience perspective and a cost-effectiveness perspective, we have a chance to prioritise renewable resources that aren’t dependent on fuel supplies and to complement that with distributed resources that equip the customer to achieve resilient outcomes and lower their costs in this decade. There will have to be a lot of spending no matter what we do, but we have a chance to make better choices.