LITTLE ROCK, ARKANSAS—One morning earlier this year, all seemed calm in the dimly lit, bunkerlike control room of the Southwest Power Pool (SPP) here, which manages an electricity grid that stretches across the high plains from North Dakota to New Mexico. Wall-size displays showed electricity flowing smoothly through SPP’s 100,000 kilometers of transmission lines, with 80% of the power coming from plants fueled by coal and natural gas.
The control room’s tranquility, however, belied an emerging, high-stakes battle to redraw this map. SPP and other U.S. grid operators are facing an unprecedented tsunami of requests from energy firms to connect thousands of proposed wind, solar, and power storage projects to their transmission lines. The projects are essential to meeting the U.S. goal of eliminating all planet-warming carbon emissions from the nation’s electricity supply by 2035, analysts say. Together, they could generate almost 2000 gigawatts of electricity—exceeding the total capacity of the country’s existing power plants.
Most of these projects, however, have been stuck in limbo for years, waiting in what energy insiders call the “interconnection queue.” One contributor to the bottleneck: mathematical simulations that SPP and other operators use to predict how electricity from those new power generators will affect the grid’s stability and reliability.
These simulations do not address the much-debated complications of relying for a steady stream of electricity on wind and solar power sources, which are subject to the whims of weather and deliver power intermittently. That remains a concern, but it doesn’t prevent new projects from coming online. Interconnection studies are focused on a different question: Will the projects generate too much power, and in the wrong places, testing the limits of what power lines can handle before they start to melt? To avoid such damaging congestion, grid operators require renewable power producers to pay up front for expensive transmission upgrades. But many can’t afford those improvements and must abandon their plans.
“Interconnection is becoming one of the leading barriers to bringing projects online,” says Joe Rand, a researcher at Lawrence Berkeley National Laboratory who tracks projects in the interconnection queue.
Some researchers and renewable power advocates argue that the interconnection logjam is, in part, a product of flawed simulations based on assumptions that are too conservative and sometimes unreasonable. “These assumptions are so important, because they ultimately dictate” what it costs to build new generation, says Aaron Vander Vorst, head of growth strategy and transmission at Enel North America, a major developer of wind and solar projects. Yet they give rise to show-stopping scenarios of overloaded lines that don’t reflect reality, he and others say.
Grid managers like SPP insist that interconnection simulations are essential to making sure new renewable power sources don’t overwhelm the grid. Yet, Vander Vorst and others argue that they have become a kind of malfunctioning toll gate on the road toward clean energy. To fix that toll gate, some want to see the adoption of more realistic model assumptions and more flexible rules. And when new power lines really are needed, they say, the burden should not fall entirely on the new green power projects.
THE INTERCONNECTION roadblock is personal for farm families who live amid the windswept fields and grasslands of eastern New Mexico. “We have the best wind in the country, you know,” says Eva Woods, a resident of Broadview, New Mexico.
A decade ago, Apex Clean Energy, a Virginia-based firm, drew up plans to harvest that power. Local farmers were ready to lease their land as sites for up to 135 wind turbines that together could generate 300 megawatts of electricity—enough to power 100,000 homes. Regular income from such leases “helps our farmers and ranchers so much,” says Woods, a well-connected figure who’s been publicly supporting the project.
In early 2017, Apex applied for permission to connect this potential project, called Grady Martin Wind, to a power line just outside the town of Clovis. The request went to SPP, which manages the region’s grid.
Grady Martin was just one of dozens of wind and solar projects that asked SPP for interconnection around the same time. Like many grid operators, SPP struggled to keep up with the surge of applications, and it took years to launch the needed study. But in 2021, SPP’s engineers were able to fire up what’s called a power flow model, which calculates the flow of electricity through every power line and transformer on the grid. They ran a simulation in which 60 wind and solar farms—including Grady Martin—replaced an equivalent amount of power from existing power plants elsewhere. This amounted to a major rerouting of electricity; the proposed projects represented about 10,000 megawatts of generation, equivalent to about 20% of the peak demand for power in SPP’s region on a hot summer day.
The model choked on the glut of new renewable energy. Its equations could not find a solution to meeting the region’s power demands with this new fleet of generators, using existing power lines. With all the wind and solar plants generating maximum power, the system couldn’t safely handle it all.
And the model hadn’t even gotten to another essential part of the study, which involved simulating what would happen if a routine problem—such as the failure of a single power line—occurred.
To relieve the congestion, SPP’s engineers had to add hefty new power lines to the simulation. Their most notable addition: what’s known as a double-circuit 765kV line running more than 200 kilometers southeast from Lubbock, Texas. Only a handful of power lines this big exist in the United States, none of them west of the Mississippi. “A 765kV line is like a 10-lane highway through Los Angeles, right?” says David Kelley, SPP’s vice president of engineering.
In the model, adding those power pipes relieved congestion and allowed all 60 wind and solar projects to connect to the grid. But there were practical hitches: Building that new 765kV line, for instance, would cost well over $1 billion, SPP estimated. And it assigned that cost to 14 proposed wind and solar farms, including Apex’s Grady Martin. Proposed projects were also expected to pay for more routine fixes. In total, the projects were on the hook for $4.6 billion in transmission upgrades. Grady Martin’s share was $272 million.
Faced with such costs, Apex withdrew its application. So did half of the 60-odd projects in the study group. The dropouts included 13 of the 14 projects that were supposed to pay for the 765kV line. When those projects evaporated, so did the need for the electricity superhighway. “That 765kV line is not getting built,” Kelley says. Nor is Grady Martin, at least for now—and farm families are getting no checks.
THE ELECTRIC utilities and independent grid managers that run interconnection simulations say they are simply following rules set by the Federal Energy Regulatory Commission (FERC), and their sole goal is to ensure a reliable supply of electricity.
FERC is now working on a revision of its rules. Renewable energy advocates, meanwhile, argue that some of the assumptions built into the simulations favor the business interests of old-style power companies, for whom the existing grid is working just fine. Those companies favor rules that protect their existing power plants from new competition, or shift the cost of valuable new transmission infrastructure to the new generators.
“There’s definitely a political aspect, and a self-interest aspect, to all of this,” Vander Vorst asserts.
Companies trying to build renewable power have won some victories. In 2021, they convinced SPP to drop an assumption that all wind and solar farms were simultaneously generating maximum power. Apart from being highly unlikely, this assumption produced simulations with so much transmission congestion that no new wind or solar proposals could even be considered in some areas. Now, SPP’s simulations assume that previously approved wind and solar plants generate anywhere from zero to 75% of their capacity, depending on the exact scenario. The change came too late for Grady Martin, but the forecasts now have fewer overloaded power lines, and wind and solar projects are saddled with fewer upgrade costs.
Critics of the models, however, want revisions to go even further. Vander Vorst, for example, says interconnection studies should consider other ways to avoid overloaded power lines, apart from building new ones.
In real life, grid managers do this routinely, easing congestion on power lines by ordering some power plants to throttle down, while increasing power generation elsewhere on the grid to compensate. If new wind or solar farms create such problems frequently, Vander Vorst says, it’s a sign that new transmission lines really are needed. But if it only happens rarely, the grid’s gatekeepers could allow new projects to interconnect and just manage those problems if they occur.
Texas already uses a version of this system. Solar and wind projects there face relatively few obstacles to interconnection. They do, however, face more financial risks once connected. If many projects connect to the grid in the same area, they can end up competing with each other for limited space on the transmission system, limiting the amount of power they can sell.
Queued up
A rapidly increasing number of solar, wind, and energy storage projects are waiting to connect to the U.S. electricity grid (top). The generating capacity of these proposed projects exceeds that of the country’s existing power plants. Many of the projects would be built in rural areas of the West and Midwest (bottom).
SPP isn’t enamored of the Texas-style approach, Kelley says. His job, he says, is to make sure power plants can deliver their power to the grid with as few constraints as possible. “The last thing we want to do is to give the real-time operators a grid that they can’t manage,” he says.
Some renewable energy companies argue for another change: letting simulations incorporate new “grid-enhancing” technologies that allow the existing grid to carry more power than is currently assumed possible.
One technology is called dynamic line rating. It automatically adjusts the amount of current that power lines are allowed to carry based on local weather. On a cool, windy day, a transmission line often can handle substantially more power without overheating than it can on a hot, calm day. Related technologies are able to redirect the flow of electricity away from congested lines toward others that have excess capacity.
FERC has proposed rules that encourage grid operators to consider grid-enhancing technologies when studying interconnection requests. But many in the power industry are skeptical.
“Some of these things are really unproven, uncommercial kinds of technologies,” Kelley says. “We can’t just put something theoretical in the model and say: ‘Oh, that makes the issue go away.’”
Dynamic line ratings, for instance, can be a kind of Band-Aid that helps control room operators resolve temporary congestion problems, Kelley says. But it’s no replacement for new power lines. “You just typically need bigger pipes in order to move more energy,” he says.
EVEN AS EXPERTS debate model tweaks that might ease clean energy connections, others are calling for more attention to the risk that wind and solar generators may react to electrical disturbances in unexpected ways. Those anomalies, they say, could pose a greater threat to the grid than congestion.
Researchers point to a near-disaster that occurred in Texas on 4 June 2022. It started with an equipment failure at a gas-burning power plant in Odessa, which caused the 550-megawatt plant to “trip,” or go offline. The shutdown perturbed the voltage of electricity flowing across western Texas.
That should not have caused serious problems. Almost instantly, however, a dozen big solar farms in west Texas, which were generating 1700 megawatts at that moment, also tripped offline. The interruption briefly threatened to take down parts of the grid.
An investigation by the North American Electric Reliability Corporation (NERC), an industry group, pointed a finger at the solar plants’ inverters—core equipment that converts the direct current from these generators into alternating current that matches what’s already flowing on the grid. The inverters’ programmable controls had been set in such a way that they didn’t “ride through” the disturbance, as they are supposed to do.
The fix, which is still being implemented, involves reprogramming some of those software controls. Yet NERC and others say the Texas incident illustrates a broader vulnerability, rooted in the fact that wind and solar generators behave differently from traditional coal or gas plants.
Those old-style plants, with their massive rotating turbines, have an inherent physical ability to ride through disturbances. Wind turbines and solar panels, by contrast, rely entirely on automated, software-driven electronic systems to control their power output. “They do what they’re told to do, and sometimes it’s not the right thing,” says Ian Hiskens, a power systems researcher at the University of Michigan.
In particular, these control systems are programmed to be “grid-following,” meaning they track the 60-hertz cycles of voltage on the grid and deliver power that matches it precisely. But that can also leave them hypersensitive to intermittent disruptions in that signal, Hiskens says. Such disruptions can be caused by events as subtle as surges in power from a nearby wind farm as the wind picks up. “You can end up with interaction between two adjacent wind or solar farms, an action-reaction kind of thing,” Hiskens says.
These problems are more likely to pop up in parts of the grid that are far from old-style generators, which help maintain steady voltage and frequency—like conductors of an orchestra keeping everyone on tempo. At a distance, that tempo signal may be indistinct and hard to follow—a situation that’s sometimes called a “weak grid.”
In Australia, for example, Hiskens says authorities have noticed intermittent voltage oscillations in a part of the grid with lots of wind and solar farms. It hasn’t been a major problem, he says, “but if you don’t understand the oscillations, then you should be pretty wary. Because oscillations tend to be precursors to instability.”
In the future, as wind and solar generators come to dominate the grid, some will likely be outfitted with “grid-forming” inverters that enable them to take over the tempo-setting role from older turbines. But that transition is still years away.
In the meantime, NERC would like to see the grid’s gatekeepers add a new kind of model to their interconnection studies. These “electromechanical transient,” or EMT, models can simulate the behavior of inverter control systems in exquisite detail, showing how they react to disturbances.
Grid operators such as SPP and the Midcontinent Independent System Operator are now carrying out such studies, but they are running into complications. The models require a lot of computing power, and people who know how to run them are in short supply. Perhaps more important, the models only deliver accurate results if they are using accurate information about the equipment that will be installed, and its control settings. That information is often unavailable, because when interconnection studies begin, solar or wind developers don’t know what equipment they’ll install years down the road.
“If you’re going to do an EMT study in phase one [of interconnection], then you’re high on crack,” says Rhonda Peters, a consultant who’s worked for many wind and solar developers. She says such studies make more sense if conducted closer to actual construction.
FERC IS CURRENTLY working on new regulations that could help ease the interconnection logjam. One set of proposed rules, released on 27 July, would impose fines on grid managers who take too long to consider interconnection requests. It also attempts to discourage renewable energy companies from clogging the queue with speculative projects that aren’t likely to succeed.
In the short term, however, the interconnection problem is likely to worsen. The Inflation Reduction Act, a 2022 law that includes hefty financial incentives for wind and solar projects, is now fueling a new surge in applications. “There’s so much activity in this industry, people are going bezonkers,” Peters says.
Over the long term, studies suggest the most cost-efficient way out of the interconnection maze is for electric utilities to figure out in advance where new power lines are most needed—and then build them. Utilities, however, are typically reluctant to impose the rate increases needed to pay for new infrastructure. “The main barrier is still, who pays? And because they cannot solve it, that’s where it all stops,” says Julia Matevosyan, chief engineer for the nonprofit Energy Systems Integration Group. A rule FERC is now developing could help by shifting the burden of paying for new power lines, so new renewable energy projects won’t have to shoulder the bulk of the cost.
For the moment, the interconnection simulations, for all their flaws, are actually revealing a fundamental truth. The U.S.’s energy transition will demand a far more ambitious expansion of electricity infrastructure than anyone, so far, has been willing to embrace.