A gate and a “No Trespassing” sign block the way for unauthorized cars to access Duke Energy’s facility on South Andrews Avenue in Hot Springs. So it takes a little legwork to see what’s been billed as the future of North Carolina’s electrical grid.

Start on the east bank of the French Broad River, across from the quiet Madison County town of about 520 people. Head south on the Appalachian Trail and hike about a half-mile up a series of wicked switchbacks.

When you reach Lovers’ Leap, a rock outcrop named for a tragic Cherokee legend, look to the west. Surrounded on all sides by the forested foothills of the Blue Ridge Mountains lies a gently sloping field of solar panels, a handful of low-slung buildings, and a few transmission lines heading toward town.

The setting may be bucolic, but Duke Energy says the roughly $14.5 million project boasts some of the most advanced electrical infrastructure deployed anywhere in its service area. Hot Springs is home to the utility’s first “inverter-only” community microgrid.

The concept of a microgrid—a set of power generation sources, storage assets, and electricity controls that can operate without being connected to a wider system—isn’t particularly revolutionary in itself. “The first grid that was out there was a microgrid. That’s what Thomas Edison built in New York,” said Robert Cox, associate director of the Energy Production and Infrastructure Center at the University of North Carolina at Charlotte.

What makes Duke Energy’s innovation different is the inverter-only piece: It exclusively uses solar panels and batteries, which rely on devices called inverters to transform their direct-current electricity into the alternating current of the power grid. The grid (and all the things it powers) is designed to operate using current at a specific frequency, Cox explained, and the physical rotation of big turbines in fossil fuel or nuclear power plants has traditionally set that frequency.

The Hot Springs project, in contrast, can form a functioning electrical grid from scratch. Digital controls allow the solar panels and batteries to mimic the rotation of a turbine, so the entire town can get synchronized to the same electrical tune without the proverbial conductor of a far-off power plant. 

The view of Duke’s microgrid project from the Lovers’ Leap overlook. (Photo by Daniel Walton)

That so-called “black start” capacity is meant to solve a recurring problem for Hot Springs. Before the microgrid was installed, all of the town’s power flowed over a roughly 10-mile distribution line that stretches through the mountains from the nearby town of Marshall. Fallen trees after high winds or heavy snowfall often interrupted service somewhere in the Doe Branch area of Madison County, a region of rugged hills and no roads. 

Duke Energy crews would have to venture into the woods on foot to fix a fallen electrical pole or break in that line, Hot Springs Mayor Abby Norton explained. Businesses could be left without electricity for hours or even days, leading to lost inventory and income.

With the microgrid, however, the town can reboot itself and stay powered for roughly four to six hours at a time as an electrical “island” while its connection to the main grid is restored. And deployed more widely, the technology Duke Energy is developing in Madison County has the potential to transform how North Carolina manages its electricity.

Adding more microgrids like that in Hot Springs could make North Carolina more resilient to extreme events like December 2022’s winter storm, which led Duke to institute rolling blackouts for the first time in state history. Those microgrids could effectively fend for themselves when electricity demand peaks, reducing strain on the rest of the grid and allowing Duke to keep the lights on, said Jason Handley, the general manager of Duke’s Distributed Energy Group.

The storage capacity and electrical flexibility of microgrids also make it easier for Duke Energy to incorporate renewable power sources. They effectively smooth out the intermittent nature of electricity from the sun and wind to ensure a stable power supply for customers. “If we can do that with inverter-only-based systems, we’re continuing the goal, which is greening our grid,” he said.

Yet those long-term advantages come with an up-front expense. Jeff Thomas, an electrical engineer with the North Carolina Utilities Commission’s Public Staff, estimates that adding microgrid capacity to an energy storage project increases its cost per kilowatt-hour by about 60 percent. And by state law, Duke Energy must practice “least-cost energy planning” when determining how to deliver reliable electricity.

Hot Springs, adds Public Staff Energy Division Director James McLawhorn, was a unique case given its remote setting and history of power outages. The utilities commission approved the project as an experiment, noting that its benefits “are material but are difficult to quantify accurately.” What Duke learns from the microgrid will help determine if similar projects can be approved through the commission’s cost-benefit analysis.

“It’s important that the public understands that the transition to a cleaner, more resilient grid does not come without costs,” McLawhorn said. “It may be cheaper in the long run, but in the short run, there is going to be a cost.”

Learning By Doing

Duke Energy has experimented with microgrids since at least 2013, when the utility began connecting solar panels and batteries at a substation in Charlotte to power a nearby fire station in the event of a grid outage. 

Over the subsequent decade, Duke continued to tinker with the technology at its Mount Holly Innovation Lab, but its only commercially deployed microgrid project had been a tiny solar and battery installation supporting a communications tower on Mount Sterling in Haywood County.

Progress on the Hot Springs microgrid itself also took place relatively slowly. The Utilities Commission approved the project in May 2019, and Duke’s subsequent announcement said the microgrid would be in service by early 2020. The facility didn’t actually enter operation until February 2023.     

“It’s important that the public understands that the transition to a cleaner, more resilient grid does not come without costs.”

James McLawhorn, North Carolina Utilities Commission’s Public Staff

Duke Energy’s Handley said part of that delay came from the utility pushing the technological boundaries of grid design. To his knowledge, no other community microgrid project in the world has achieved black-start capacity using only solar and battery power. (A similar project in rural Australia can power an entire town using renewables alone, but isn’t able to restore service independently after an outage.)

In Hot Springs, Duke Energy engineers had to make sure the microgrid would be safe and stable when operated as its own island. That required much more difficult calculations than what’s required to join a system to the larger grid. 

Duke Energy’s Junior Hatcher, who led the team that built the Hot Springs microgrid. (Photo courtesy of Duke Energy)

“To be quite honest, that’s PhD-level work. We found that we had to contract this out to some other companies that had experience doing this on transmission systems,” Handley said. “This setup is our blueprint for everything else we’re doing [on community microgrids], so we had to ensure we got it right.”

The project also experienced some of the supply-chain challenges that plagued many industries throughout the COVID-19 pandemic. At one point, Duke faced yearlong lead times on certain transformer components. And even following extended preparations, the Hot Springs facility ran into trouble the first time it faced the type of outage it had been designed for.

After high winds knocked out main grid power to the town on April 1, the microgrid’s automated start-up system didn’t work as designed; Duke had to send in technicians from Charlotte, a nearly three-hour drive away, to reset the electronics and restore power. The utility never held an official ribbon-cutting and media visit that had been floated for the site.

Learning by doing—and sometimes failing—is critical to bringing more microgrids online, Handley stressed. “We’ve given ourselves a 75 percent to 85 percent blueprint for how to do the next one,” he said. “Microgrids will never be cookie-cutter, but what we’re trying to do is get them scalable and repeatable. That is what we’ve got to be able to do to make these something that we can offer our customers for the right price and on the right project scale.”

The next wave of these projects is already in motion. At Camp Lejeune, the Marine Corps training center in Jacksonville, the utility is building a $22 million microgrid with black-start capability like Hot Springs, although the facility will also include natural gas generation. Handley says similar projects are planned for Camp Atterbury-Muscatauck, a National Guard facility in Indiana, and several locations in Florida.

For now, Duke is targeting microgrids for places where reliable power is particularly important. Handley points to a project planned for a middle school in St. Petersburg, Florida, which doubles as a hurricane evacuation shelter for people with special needs. But he says the utility isn’t limiting itself as it makes grid upgrades.

“My team has been tasked with really taking any project that we get and seeing if it can be turned into a microgrid,” Handley said. “We believe that, in the long run, that’s going to make the best outcomes for our customers from a cost and reliability standpoint.”

Light Speed?

In the short run, however, Duke Energy doesn’t seem to believe widespread microgrids are a viable approach to upgrading the grid. 

Of the more than 6,700 “non-traditional” electricity distribution and transmission projects—essentially, any alternative to big lines from centralized power plants—the utility included in its most recent Carbon Plan and Integrated Resource Plan filing, only one was deemed both technically and economically feasible for North Carolina.

Battery storage at the Hot Springs microgrid site. (Photo courtesy of Duke Energy)

On the power generation side, smaller-scale renewables like the solar panels deployed in Hot Springs appear to be the exception rather than the rule. Duke’s filing projects that at least 68 percent of North Carolina’s power in 2050 will be provided by nuclear sources, up from 46 percent today; solar will yield no more than 22 percent, up from 6 percent currently. Duke said that in the future, it “will primarily procure substantially larger, transmission-connected” solar projects.

Duke Energy’s plans represent a massive missed opportunity, said Walt Bussells, the former CEO of Florida’s largest community-owned electric utility. He is also now a volunteer for the Asheville-based Critical Services Microgrid Group, which works to promote advanced electrical infrastructure in the region.    

“The legacy grid, lots of wires and gigantic power plants a long way from where people live, has served our country extremely well for 120 years,” Bussells said. “But current technologies and the economics of things say its time has passed. There are so many opportunities now to deploy clearly superior approaches to providing essential services that are decentralized, precise, clean, renewable, and cheaper.”

Bussells believes that Duke’s cost-benefit analysis for microgrids doesn’t sufficiently value their auxiliary benefits, such as reliability in a world of increasing climatic and social instability. He points to Babcock Ranch, a Florida community of about 5,000 that stayed electrified throughout the Category 5 Hurricane Ian in 2022, thanks to its solar-powered microgrid. 

“Microgrids will never be cookie-cutter, but what we’re trying to do is get them scalable and repeatable.”

Jason Handley, general manager of Duke’s Distributed Energy Group

He also points to the December attack on two substations in Moore County, which left tens of thousands of people without power for days. “With a system of decentralized microgrids, a single attack might take out 50 houses or two buildings, but not an entire county like it would with the legacy grid,” Bussells said.

But regulators must be convinced of the benefits of microgrid projects to approve them, notes a report from North Carolina State University’s Clean Energy Technology Center, and it can be difficult to define and determine those benefits. 

Handley, the Duke Energy engineer, said that the real-world experience of operating the Hot Springs microgrid will help quantify some of those benefits. If Duke can put a dollar value on intangible goods such as avoided downtime and smoother electrical service, the utility can make a stronger case for distributed grid infrastructure to state regulators.

Matt Abele, interim executive director of the N.C. Sustainable Energy Association, argues that the state’s regulated electricity market, which gives an effective monopoly to Duke across most of the state, locks out smaller players who would be willing to develop microgrids in response to consumer demand for resilient, renewable power.

“We think there’s a real onus for the state to study and investigate a more deregulated market structure to provide more customer choice,” he said. “That would then allow customers to be able to have more say in what the grid makeup looks like.”

The General Assembly has shown at least some willingness to explore deregulation as a way to accelerate grid updates. House Bill 503, filed in March and billed as a “Storm Resilience Study,” would have awarded $500,000 to the University of North Carolina Collaboratory to examine electricity market reform.

“It is critical that we take a close look at our current electricity delivery system during a time of impending change in both power generation and electricity usage,” said state Rep. Eric Ager, a Democrat from Buncombe County who co-sponsored the proposal.

However, Duke Energy strongly opposed the bill. Although over 20 representatives from both sides of the aisle had signed on as co-sponsors, it did not make it out of the House Rules committee.

As engineers continue to solve the technical challenges of microgrids, Bussells suggests, it’s those types of very human disagreements that will prove the biggest obstacle to a better power system. 

“It’s the madness of crowds, as they say. How we behave in larger organizations sometimes is irrational,” he said. “That feels like the barrier.”

Daniel Walton is an Asheville-based freelance reporter covering science, sustainability, and political news. He was previously the news editor of Mountain Xpress and has written for The Guardian, Civil Eats, and Sierra. Contact him at danielwwalton@live.com.