(Ed: A well written article describing some of the software requirements for a VPP that addresses EV charging.)
Electric vehicles could overwhelm local grids. Managed charging offers a solution, and utilities are eager to get their EV-driving customers signed up.
By Jeff St. John
13 December 2024
Millions of electric vehicles are going to plug into U.S. power grids over the next decade. The coming wave of EVs could wind up overloading the grid and costing utilities and their customers a lot of money — or it could be a smooth integration that leads to savings for ratepayers.
It all depends on how well utilities plan for and handle EV charging.
Some of the earliest work on EV-charging management is happening in the country’s hottest EV market — the San Francisco Bay Area, or specifically, Alameda, Contra Costa, and Santa Clara counties, where EVs now make up nearly half of all new car sales.
It’s in these counties that Pacific Gas & Electric is trying out an innovative program to make EVs play nicer with the grid: In October, PG&E launched its EV Charge Manager program. The goal is to enlist about 7,500 customers willing to earn up-front and ongoing incentives in exchange for letting the utility control when their EVs charge from hour to hour.
Charge Manager aims to test what Lydia Krefta, PG&E’s director for clean energy transportation, described as “active managed charging.” That’s different from the time-of-use rates and other more common utility programs, which encourage EV owners en masse to avoid charging during certain hours when regionwide electricity demand spikes and start charging later at night when electricity demand eases.
Instead of turning all the EVs on and off at the same time like time-of-use rates and other similar programs aim to do, managed charging modulates how EVs power up throughout the day and night in response to how they’re actually affecting the local grids they’re connected to.
To do that, utilities will have to predict and manage “hot spots” for EV adoption. They’ll have to convince customers to charge EVs in ways that reduce the need for costly grid upgrades. And they’ll have to convince regulators that the technology needed to enable this “smart” charging behavior is a sound investment.
“We want to validate this hypothesis that we can not only manage EVs in ways that help reduce overall transmission and generation needs,” Krefta said, but also “in a way that helps defer upgrades that are needed on circuits and transformers on the distribution system.”
Why the distribution grid needs attention
Utilities don’t monitor and control distribution systems — the lower-voltage grids that deliver power to individual customers — as closely as they do transmission systems that ferry electricity from power plants to big substations.
That makes it harder to optimize charging among multiple customers sharing the same transformers and circuits on the distribution grid — or to predict the benefit from shifting charging around.
That’s a problem. These distribution assets will start to struggle to meet EV-charging demand well before the broader utility systems will. The scale of the problem is hard to predict given the lack of visibility into distribution grids, but it’s unlikely to be small: Studies of California’s three major utilities, including PG&E, show that upgrading distribution grids to handle the 13.5 million EVs expected to be in the state by 2035 could add up to more than $50 billion.
Managed-charging programs like PG&E’s new pilot might offer a way to keep those costs in check, cutting in half the price of a distribution grid upgrade through 2035, according to more recent studies of California’s grid.
Waiting for millions of EVs to get plugged in before taking action won’t work, said Kyle Garton, head of product for WeaveGrid. The startup, which raised $28 million this week, is involved in similar programs with utilities including Baltimore Gas & Electric in Maryland, Xcel Energy in Colorado, and Portland General Electric in Oregon.
That means utilities must act now for managed charging to save customers money, as EV adoption is set to rise fast. “We have a 2030 goal to support 3 million EVs — we’re at about 650,000 as of mid-2024,” Krefta said.
Grid problems can emerge even at relatively low levels of EV adoption, Garton said. Certain neighborhoods — like those in the fancier parts of Silicon Valley — are going to end up as “pockets of higher adoption, and they can grow quickly and unexpectedly,” he said.
Finding the ‘tipping point’ for EVs to affect the grid
EV hot spots aren’t just a problem for states with aggressive climate mandates like California and New York.
A recently released study sponsored by the Smart Electric Power Alliance (SEPA) trade group, featuring data from WeaveGrid, Toyota, and EnergyHub, a company working with dozens of utilities on programs that control customer thermostats, appliances, and EV chargers to reduce grid costs, found that distribution grid upgrades driven by current EV adoption trends “are estimated to cost billions of dollars and could average more than $2,500 per utility customer.”
Another recent study from energy and utility company AES Corp. discovered similar dynamics facing AES Indiana, a utility serving about 500,000 customers in the Indianapolis area — not exactly a hotbed of EV adoption. That report found that the cost of implementing managed charging started to yield benefits when EV penetration across the utility hit 5 percent, which AES Indiana expects to happen around 2029.
That’s because “you don’t just see people randomly adopting EVs,” said Astrid Atkinson, CEO and co-founder of Camus Energy, a startup providing cloud-based software to manage utility grids that worked with AES on the study.
Instead, there’s a “clustering,” she said, with certain neighborhoods quickly reaching a concentration of EVs that can cause distribution grid overloads. Managed charging for those clusters could save a lot of money for AES Indiana customers at large, the report found.
That’s because the practice could allow the utility to defer grid upgrades at those spots, potentially for several years. The report found that the value of deferral between 2025 and 2035 was $75 million per year, or $1,275 per year for every new EV — a lot more than what any utility now pays EV owners to participate in managed-charging programs.
About 75 percent of those savings are at the distribution grid level, said Colleen Lueken, director of market development for AES’s Organic Grid initiative, and only 25 percent from reducing system peaks that drive transmission grid and power plant investments. Unfortunately, “often that 25 percent is all that’s accounted for in looking at these benefits,” Lueken said.
In fact, some solutions for those systemwide problems can magnify local challenges. Time-of-use rates can actually worsen wear and tear on distribution grids, due to what’s known as the “snapback effect” — the phenomenon of EV chargers restarting en masse when high prices end and cheaper rates start.
That can “inadvertently create new peaks on local equipment,” which “drive increased capital expenditures compared to no EV management,” the report found.
Distribution-system benefits and costs usually aren’t calculated because utilities don’t have enough data to understand them, Atkinson said. At AES Indiana, Camus tapped into existing smart meter data to extrapolate the impact of adding EVs to homes and correlated that with AES’s systems to track the status of its distribution grid assets.
It then modeled how the extra EV-charging load would affect those assets to predict how many more years they could last through managed charging. Without that data, it would be hard for AES Indiana to prove to its regulators that the cost of deploying targeted managed charging would yield benefits that make it worthwhile.
AES Indiana hasn’t yet asked state regulators for permission to track real-time conditions of its distribution grid assets, Lueken said. But its analysis included the cost of tracking technology in its calculations that showed how much managed charging could save the utility.
“It’s really about asking, where do you hit a point — what we call a tipping point — where investing in that visibility and coordination is a more prudent course of action than doing upgrades,” she said.
Some utilities are trying more advanced technology to get this data. In October, Consumers Energy, which serves about 2 million customers in Michigan, won a $20 million Department of Energy grant to deploy 20,000 next-generation smart meters developed by grid technology vendor Utilidata in partnership with chipmaker Nvidia.
The custom chips inside these specialty smart meters provide “a level of real-time visibility into what’s happening on the system, and where EVs are charging and the impact of that, that’s leaps and bounds beyond what they’ve had before,” according to Utilidata President and COO Jess Melanson.
Michigan had only about 50,000 EVs on its roads as of September, and Consumers Energy hasn’t yet faced distribution grid challenges from EV charging, said Ryan Jackson, the utility’s executive director of corporate strategy and strategic projects. Governor Gretchen Whitmer, a Democrat, has set a goal for 2 million EVs in the state by 2030.
“We’re very successful in deploying time-of-use rates today, with a high rate of adoption, and we’re seeing the benefits,” he said.
Still, Consumers Energy was unsure if TOU rates are “enough to mitigate deep penetration of EVs,” Jackson said, or whether they needed “a more coordinated effort to avoid a local peak, where everyone starts charging at 2 a.m.”
That could help Consumers figure out where grid constraints might eventually present challenges for customers who want to install EV charging at their homes, he said. In that sense, “we’re solving a future problem so we don’t have to tell that next customer, ‘You can’t charge your vehicle.’”
A similar pilot project launched last month by PG&E and Itron, a smart meter and grid technology company, is aimed at solving a similar problem. It’s targeting up to 1,000 customers whose electrical panels are too small to support higher-powered home EV chargers — at least at the same time that other big household electrical loads like air conditioners or washer-dryers are also running.
PG&E will give those customers one of Itron’s latest-generation smart meters, which can connect via WiFi to monitor and control EV chargers and other energy-sensing devices in people’s homes, said Quinn Nakayama, PG&E’s senior director of grid research innovation and development.
“Your EV doesn’t have to charge at your maximum charge rate when your air conditioner is running or your washer-dryer is running,” he said. Upgrading electrical panels can cost thousands of dollars and take weeks or months, giving customers an incentive to choose a load-control alternative.
Those Itron meters can also inform PG&E of risks to its grid, Nakayama noted. “If you’re in the middle of September, everyone’s running their AC systems and charging their EV at the same time, we’re going to overload that service transformer and have to upgrade it,” he said. “We can tell your EV charger not to charge when everyone’s running their AC — or maybe charge it at a lower rate.”
To be clear, PG&E isn’t proposing to make this a standard response to customers’ at-home EV charging, Nakayama said. “This is a learn-by-doing exercise. We want to work with these customers.”
If anything, Krefta said, utilities have strong incentives to use as light a touch as possible in shifting when customers charge their EVs. “Given the adoption curve for EVs today, and the fact that we want this to be widespread, customer experience is really important,” she said. “We want customers to welcome this.”
Luckily, most EVs need only a few hours of charging at home to fully replenish their batteries. That gives managed-charging programs plenty of overnight hours to move charging sessions around, as long as they ensure that EVs have full batteries by morning.
PG&E is also trying to figure out how much to pay participants. Every customer who enrolls in Charge Manager will get a $75 gift card, with more incentives for those who stick with the program for at least three months. But PG&E makes clear on its signup page that it’s offering “varying incentive amounts to participants,” that is, some people will get more than others.
“We’re trying to use this pilot to understand response rates to different incentives,” Krefta said. “That’s one of the open questions. If I save $10, is that worth it to me? Is $100 worth it? What does it take to motivate a customer? You don’t want to compensate more than the value it’s creating.”
On the other hand, telling people they can’t plug in EVs until utilities expand their grids isn’t a good option either, she said. Connecting as many EVs as possible isn’t only critical to clean transportation goals and giving customers the service they expect — it also drives down utility rates.
Utilities recover the costs of building grid infrastructure from their customers. As long as the new revenues that utilities collect from selling power to EV owners exceeds the incremental costs of building the grids to serve them, EVs help drive down rates. That’s been the case in states across the country over the past decade, according to studies from Synapse Energy Economics.
But to ensure that EVs remain a cost-reducer for utility customers at large, those grid costs must be kept in check. To make that happen, utilities have to install managed-charging programs — and prove that they work — in time to prevent grid overloads from happening, which would force utilities to react in costly ways to fix them.
Over the next year or two, PG&E will collect data on how controlling charging schedules is altering the stress on transformers, feeder lines, and other equipment, Krefta said. The utility already has smart meters, distribution grid sensors, and other technologies that can provide data to determine the relative balance of grid costs and upgrade deferral benefits.
That will answer questions like, “what’s more cost-effective, upgrading the transformer, or managing vehicles at that transformer constraint?” Krefta said. “That allows us to compensate the customer for participating, but also creates some value for all PG&E customers at the same time.”
Those findings, in turn, will help inform PG&E’s future rounds of distribution grid upgrades, Krefta said. “We’re trying to nail down how much more we can count on this to build into our planning.”
“Ideally, all goes well, and we decide we would want to scale this more broadly to EVs beyond those three counties,” she added.