Grid reslience is everyone's problem

So is climate change

Hurricane Helene devastated southern Appalachia over the weekend. There are people who are better than me at directing how to help, but if you want to take my advice, I will send you to the Foster Family Alliance of North Carolina. They are on the ground in the area and helping foster families relocate to hotels and Airbnbs to get them to a secure location. 

The impacts of Helene are going to be felt for months if not years in an area that would have never expected to see this scale of impact from a hurricane system. As recently as 2018 when Hurricane Michael became just the fourth category 5 hurricane to make landfall in the US, the wreckage was limited to as far north as central Georgia despite similar pathing to Helene. For that storm, Atlanta experienced wind gusts of just 20 mph for Michael whereas Helene generated hurricane warnings with gusts up to 70 mph for Atlanta. There are many factors that determine how a hurricane performs but undoubtedly this storm is a harbinger for the future of extreme weather worsening due to atmospheric warming.

With all of this in mind and acknowledging that natural disasters impact much more than the electrical grid, let’s talk about the gaps in future energy security. Fair warning: this one gets pretty ranty.

Anomalies versus infrastructure

Throughout the history of electricity usage, the assumption has always been that it is better to build a bigger generator to maximize the capital investment. Making the turbines that turn to generate the electricity as big as possible sets up a successful future where fuel supply lines know where they have to go and electricity can be transmitted hundreds of miles if necessary to keep communities from having to look at a smokestack.

This method leaves behind anomalies in rural communities that are hard to reach with power lines. That’s why where I grew up in Keeseville, NY, many families had diesel or gasoline generators that would power their essential appliances when the power inevitably went out. The cost of unreliability was passed to the consumer. Adding decentralized generation to the old centralized generation model doesn’t mean we have to retire all centralized resources right away. It does mean that we add resilience to the grid incrementally and have the benefit that the decentralized model is powered most cost-effectively by clean energy resources.

Asheville, NC is not a rural community. It is a city with an urban area containing 400,000 people that happens to be surrounded by mountains. Greenville, SC is slightly less mountainous but more than double the size with a metro area containing more than 900,000 people. Both cities have significant power outages even after 3 days due to flooding. In the Asheville area, more than 360 substations have been damaged, many of which were submerged, and it remains unclear if they will need to be completely replaced. Even power lines that were buried to avoid weather impacts have been severed in landslides. In the urban scenario where resiliency is expected to be the best, it has failed these two mid-sized cities. Adding rooftop solar and batteries to most buildings in the city would give folks at least the opportunity to access critical services from luckier neighbors or businesses who did not have their homes washed away.

Most concerning is that FEMA lists Asheville, NC as one of the least risky areas in the US but now the climate is changing. The high risk is no longer to the person who chose to build a cottage on the outer banks of NC that could be swept away with a storm at high tide. Building a grid that is resilient to new weather means decentralization. Failures should be assumed to happen constantly and the goal is to limit the impact to avoid catastrophe. Unfortunately, the science points to this becoming more commonplace, especially in mountain areas. Asheville and Greenville are typically powered by as much nuclear power as anywhere in the US, but those plants can’t help anyone if they aren’t able to send their power. That centralized model is why many of the nuclear power plants (all seemingly unaffected by the weather) have powered-down reactors with nowhere to send their power.

Pilot programs that are too slow

Here in NC, there are two pilot programs in place to test the idea of decentralization in order to build a more resilient grid. 

First, is the EnergyWise program that I am a part of. Duke Energy essentially pays for the battery portion of my rooftop solar plus battery installation over 10 years in exchange for being able to control the battery in times of grid stress. The program is limited to only new solar installs and I’m part of a group of 148 homes. 

Second, is a community microgrid pilot in Hot Springs, NC that isn’t so much a microgrid but a large solar plus storage array that powers a town of 500-600 people. The innovation is that Duke Energy installed a switch where the 10-mile feeder line enters the town so that when that line is cut, the community can still run on solar plus storage. This means that the town residents don’t need to worry about weekly power outages in the winter, but it is worth noting that 3 days after Helene hit, Hot Springs is still showing 300 customers without power on their outage map.

These two pilot programs are good ideas but are not ambitious enough to be called pilot programs. A pilot should be something challenging boundaries, these are just solar plus storage solutions that are being mostly paid for by customers to the benefit of grid reliability. These technologies are already proven at scale in multiple places in the US and around the world. The only hypothesis being tested by these programs is Duke Energy’s revenue model at the expense of grid reliability and small business owners who could be installing solar and storage across the state if not for a series of confounding permitting issues.

Downstream deployment hurdles

Building a more distributed grid requires more sophisticated grid technologies that we already know how to build in the US. There are hurdles to visibility into distribution networks and where the money going to come from for expensive endeavors like reconductoring. However, the most critical hurdle is the ongoing shortage of transformers that will be made worse by natural disasters. Building a new substation requires many high voltage transformers and wait times for these have gone from a few months to up to 5 years since 2020. If 360 substations can be dropped in one storm, they need to be replaced and will take priority over new power buildout because people are actively suffering without power in those areas.

The federal government has provided money and opportunity, including here in NC, to companies to address the backlog. However, if we are to believe that artificial intelligence is going to be both a revolutionary profit machine and a new driver of electricity growth in the US, then it is time to ask the wealthy companies that are stressing the grid to pitch in as well. Power purchase agreements like the one that Microsoft signed to repower Three Mile Island should become standard practice for companies who want to build new data centers. Requiring that new energy procurement must be clean and absorb the cost of transmission would shift the value back toward the electrical infrastructure providers that need a market incentive to build a more resilient supply chain for our energy infrastructure in the US. The market has atrophied after more than twenty years without growth. It should’ve been fixed years ago but the next best time to start is now.