Virtual Power Plants: The Grid Capacity Nobody Had To Build |

Click Here for 150+ Global Oil Prices

Kinder Morgan Restores U.S.-Mexico Gas Flows After Pipeline Outage

WTI is caught between recovering…

China dominates the global battery…

Charles is a writer for Oilprice.com

On the evening of Sept. 9, 2026, with a heat wave pushing electricity demand across California toward its limits, more than 140,000 home batteries started discharging at the same time. Over a three-hour window, they sent more than 580 megawatts of power into the state's grid, roughly what a large conventional power plant puts out. Nobody broke ground on anything and no fuel got burned. The batteries were already mounted on garage walls across the state, and software told them when to go.

That's a virtual power plant, and it's quickly becoming one of the grid's most useful tools. Power demand is climbing again thanks to AI data centers, electric vehicles and heat pumps, while shocks like the 2026 Strait of Hormuz crisis keep reminding governments how exposed fuel-dependent power systems can be. Grid operators need more capacity, and they need it faster than new power plants and transmission lines can be built.

This guide explains what a virtual power plant is, how one works, what VPPs are actually good for and where in the world they're being deployed at scale.

Virtual power plant, defined: A virtual power plant (VPP) is a network of small, distributed energy resources, such as home batteries, smart thermostats, electric vehicle chargers, rooftop solar and flexible commercial equipment, that software coordinates so they act like a single power plant. When the grid needs help, a VPP can push stored electricity onto the grid, cut demand from it, or both.

A conventional power plant is one big machine in one place. It burns gas or coal, splits atoms or spins a turbine, then ships electricity out over transmission lines to everyone else.

A virtual power plant works the other way around. Instead of one large source, it pulls together thousands, sometimes hundreds of thousands, of small resources that already exist on the grid: a battery in a garage in Fresno, a smart thermostat in Houston, a cold-storage warehouse in New Jersey that can idle its compressors for an hour. On their own, none of those matter much to a grid operator. Bundled together and controlled as one resource, they can deliver hundreds of megawatts on command.

The word "virtual" throws a lot of people off. The electricity is real. What's virtual is the plant, since there's no single site, no smokestack and no fence line, just a lot of hardware spread across a lot of addresses.

The U.S. Department of Energy describes VPPs as aggregations of distributed energy resources that can balance supply and demand and provide grid services the way a traditional power plant does. That last part is what separates a VPP from a neighborhood full of energy-efficient gadgets. A VPP is something a grid operator can dispatch, count on and pay for.

Every VPP has three layers working together.

Here's what a typical event looks like. A utility expects demand to spike around 6 p.m. on a hot weekday, right when solar output is fading and air conditioners are running hard. The aggregator gets notice a day or a few hours ahead. During the afternoon it might precool enrolled homes by a couple of degrees and make sure batteries are topped off with midday solar. When the event window opens, batteries discharge, thermostats ease off, and water heaters and EV chargers pause. When it closes, everything goes back to normal.

Participants get paid through bill credits, per-event payments, upfront incentives or better tariffs, depending on the program. Most of them barely notice an event happened, which is the whole idea.

Almost anything that uses, stores or generates electricity and can be controlled remotely can be part of a VPP. These are the most common building blocks:

Discharge stored power during peaks and respond to grid signals in seconds

Tesla Powerwall and Sunrun fleets in California

Precool or ease off heating and cooling to cut demand for a few hours

Renew Home's network of more than 8 million connected devices

Shift heating to off-peak hours, acting like a thermal battery

Delay or slow charging, and with vehicle-to-grid setups, send power back

Smart EV tariffs managed by Kraken in the U.K.

Rooftop solar and smart inverters

Supply power and help manage local voltage

Australia's VPP-capable inverter rules for subsidized batteries

Commercial and industrial loads

Trim usage at factories, cold storage, offices and data centers on request

Voltus and other industrial aggregators

Pool solar farms, wind turbines, biogas and small hydro for trading and grid services

Broadly, VPPs come in two flavors. Some mostly cut demand, like thermostat and industrial load programs. Others add supply, like battery fleets or pooled small generators. Newer VPPs increasingly do both. In North America, the market is still made up mostly of thermostats and commercial and industrial demand response, though home batteries are the part grabbing headlines.

VPPs get lumped in with demand response a lot, and the overlap is real. Demand response, where big customers agree to cut usage when the grid is tight, has been around for decades and still makes up much of what gets counted as VPP capacity today. A modern VPP goes further by adding batteries and other resources that can push power back onto the grid and respond automatically, often within seconds.

Cuts demand, adds supply and provides fast grid services

Typically months, using devices already installed

Utility or independent power producer

Customers, installers and aggregators

VPPs aren't a replacement for every power plant. They're good at specific jobs, and those jobs happen to be some of the most expensive ones on the grid.

Peaker plants, usually gas-fired, sit idle most of the year and fire up during the handful of hours when demand is highest, which makes the power they produce expensive. DOE analysis found that a VPP made up of residential thermostats, water heaters, EV chargers and home batteries could provide peaking capacity at roughly half the net cost of a utility-scale battery or a new gas peaker. The department estimates that tripling U.S. VPP capacity to 80 to 160 gigawatts by 2030 could save about $10 billion a year in grid costs.

A new gas plant can take years to permit, finance, interconnect and build. A VPP uses devices that are already connected to the grid, so the bottleneck is enrollment and software rather than steel and concrete. That speed matters. DOE expects U.S. peak demand to rise from about 800 gigawatts in 2024 to roughly 900 gigawatts by 2030, driven by data centers, manufacturing and electrification. In Texas, Base Power, which installs home batteries and runs them as a fleet in ERCOT's market, told the grid operator in August 2026 that it was adding about 35 megawatts of new capacity a month.

Solar floods the grid at midday and fades right as evening demand peaks. VPPs help close that gap by soaking up cheap surplus power when it's abundant and releasing it, or cutting demand, when it's scarce. In the U.K., Kraken shifts power use for hundreds of thousands of EVs and heat pumps to the hours when electricity is cleanest and cheapest. Energy storage of every size is fast becoming the main tool for smoothing out that mismatch.

Batteries can respond in fractions of a second, faster than any spinning turbine, which makes VPPs useful for frequency and voltage support, not just bulk energy. South Australia's VPP, built largely on social housing, helped keep the lights on when Queensland's Kogan Creek coal unit, the biggest in the country, tripped offline and system frequency dropped well below normal.

Because VPP resources sit where electricity is actually used, they can relieve congestion on neighborhood circuits and substations and push back expensive upgrades. After the September dispatch in California, Sunrun CEO Mary Powell said distributed batteries can "reduce the need for new costly poles and wires," which is a big deal for utility customers who ultimately pay for those upgrades.

This is the newest use case, and the one moving the most money. Hyperscalers need grid capacity faster than utilities can build it, and some are now paying for VPPs to free it up. Google signed a three-year deal with Voltus in June 2026 to fund up to 100 megawatts a year of VPP capacity in the PJM market, paying local homes and businesses to be flexible instead of waiting on new generation. That same month, Sunrun, Tesla and Renew Home announced a framework for up to 16.8 gigawatts of flexible capacity, roughly 7.8 gigawatts from home batteries and more than 9 gigawatts from smart thermostats, aimed at data centers and utilities.

Households get paid for something their equipment was mostly going to do anyway. In Australia, the Clean Energy Council has pointed to analysis from the country's competition regulator showing that customers who sign up for VPPs are paying the lowest power bills in the country.

VPPs exist in dozens of countries, but a handful of markets account for most of the action, and each got there for different reasons.

About 37.5 GW in North America; DOE targets 80 to 160 GW by 2030

California's DSGS and ELRP programs, ERCOT's ADER pilot, Sunrun, Tesla, Voltus, Uplight

More than 500,000 home batteries installed under Cheaper Home Batteries, which must use VPP-capable inverters

AGL, Tesla, state VPP incentives

Next Kraftwerke aggregates more than 10 GW across eight countries

Next Kraftwerke and sonnen, both owned by Shell

Kraken controls about 5.7 GW, including 2.5 GW of residential resources

Targets of 20 GW by 2027 and 50 GW by 2030

NDRC, NEA, provincial grid companies

Balancing market opened to household batteries in April 2026

TEPCO, Tokyo Gas, Osaka Gas, Kyocera

The U.S. is the biggest VPP market by capacity. Wood Mackenzie's latest market report puts North American VPP capacity at 37.5 gigawatts, up 13.7% in a year, with California, Texas, New York and Massachusetts accounting for 37% of deployments. Most of that is still commercial and industrial demand response and thermostats. Residential customers make up just over 10% of VPP capacity in wholesale markets.

California is the showcase. The state runs two programs, the California Energy Commission's Demand Side Grid Support program and the California Public Utilities Commission's Emergency Load Reduction Program, that pay home batteries and other resources to perform when the grid is stressed. Those are the programs behind the Sept. 9 record. Texas is catching up through ERCOT's Aggregate Distributed Energy Resource pilot, which raised its participation cap from 200 megawatts to 500 megawatts in March 2026 as enrollment grew. In Utah, Rocky Mountain Power's Wattsmart program is one DOE has singled out as among the most advanced VPPs in the country.

Federal planners want a lot more. DOE's VPP Liftoff report calls for 80 to 160 gigawatts of VPPs by 2030, enough to serve 10% to 20% of peak load, and says deployment needs to speed up to get there. For more on the U.S. buildout, see Oilprice's look at how VPPs fit into America's energy transition.

Australia has more rooftop solar per person than almost anywhere, which made it an early VPP testbed. In 2018, Tesla and the South Australian government unveiled a plan to put solar panels and batteries on as many as 50,000 homes and run them as one network. AGL Energy bought that South Australia VPP from Tesla in 2025, taking over about 7,000 Powerwalls.

The bigger story now is federal. Since the Cheaper Home Batteries program launched on July 1, 2025, Australians have installed more than 500,000 subsidized home batteries. The government expects the program to support more than 2 million batteries and about 40 gigawatt-hours of storage by 2030. Grid-connected batteries under the program have to sit behind a VPP-capable inverter, but actually joining a VPP is still optional, and getting more of that fleet enrolled is the industry's next big push.

Europe's VPPs grew up differently. Instead of starting with home batteries, they started with small generators. Germany's renewable energy rules push many wind, solar and biogas operators to sell their power on the wholesale market, and aggregators sprang up to do it for them. The biggest is Cologne-based Next Kraftwerke, which passed 10,000 megawatts of aggregated capacity in 2022 across eight European countries, pooling solar, wind, hydro and bioenergy plants with flexible consumers and storage. Shell bought the company in 2021 and also owns German home battery maker sonnen, which gives one of the world's biggest oil companies a real stake in the VPP business.

In the U.K., the biggest player is Kraken, the software platform that started inside Octopus Energy and now operates independently. Kraken said in December 2025 that it controls more than 5.7 gigawatts of flexible capacity, including 2.5 gigawatts of residential resources like EVs and heat pumps, which is more than Poland's Belchatow Power Station, Europe's largest coal plant. EV owners on smart tariffs make up a big chunk of that, since their cars can charge whenever power is cheapest.

China came to VPPs later but is scaling quickly now that Beijing has set targets. In April 2025, the National Development and Reform Commission and the National Energy Administration issued guidelines calling for 20 gigawatts of VPP capacity by 2027 and 50 gigawatts by 2030. Local grid companies have been racing to build VPP platforms since, and Huatai Securities expects China's VPP market to grow from about 10.2 billion yuan in 2025 to more than 100 billion yuan by 2030.

Japan has run government-backed VPP trials since the late 2010s, with utilities, electronics makers and telecom companies all involved. The big change came in April 2026, when the country opened its balancing market to household-level resources, including home batteries. Utilities and gas companies like TEPCO, Tokyo Gas and Osaka Gas are now signing up customers' batteries, often paired with storage systems from Kyocera and Sharp.

The VPP business pulls in companies from all over the energy world. Hardware makers and installers like Tesla and Sunrun own or manage huge fleets of home batteries. Software companies and aggregators like Voltus, Uplight, Renew Home and Kraken handle the orchestration. Voltus says it manages 8.1 gigawatts of distributed capacity, while Uplight reports 8.5 gigawatts of dispatchable capacity under management.

Utilities and retailers want in, too. AGL is targeting 1.6 gigawatts of decentralized assets under orchestration by its 2027 fiscal year. And the buyers are changing. Grid operators and utilities used to be the only customers, but tech companies building data centers are now paying for VPP capacity directly, and even oil majors have bought their way into the space.

VPPs are growing, but not as fast as their backers would like. In its latest report, Wood Mackenzie found the number of VPP deployments grew more than 33% while total capacity grew just 13.7%, and it pinned the gap on utility program caps, changes to how capacity gets accredited and barriers in wholesale markets.

Enrollment is the other big hurdle. Small customers run into data-access and paperwork problems, and most people simply aren't thinking about the grid. "The biggest thing is that customers don't care about virtual power plants," Sunrun's Chris Rauscher told Renewable Energy World. Australia shows that even when batteries are VPP-ready, owners don't automatically sign up. Hundreds of thousands of subsidized batteries have gone in, but VPP enrollment has lagged well behind.

There are physical limits, too. Most VPP resources can only deliver for a few hours at a time, since batteries run down and people won't tolerate a warm house all evening. That makes VPPs great for peaks and fast grid services but not a substitute for power that runs around the clock. And coordinating them isn't simple. As Rice University professor Daniel Cohan put it, "It's a lot harder to manage millions of consumers than dozens of power plants."

Still, the basic math keeps getting better. Homes and businesses are installing batteries, smart thermostats and EV chargers by the millions, and every one that joins a VPP is capacity nobody had to build.

Not exactly. Demand response, where customers cut usage when the grid is tight, is one of the things a VPP does, and older demand response programs make up a large share of today's VPP capacity. A VPP can also add power to the grid through batteries or small generators, and it's usually automated and coordinated by software in real time.

In many places, yes. Utilities, battery manufacturers, solar installers and electricity retailers all run VPP programs, and enrollment often happens right in the battery's app. What's available, and what it pays, depends on where you live and who your utility is.

Programs typically keep part of your battery in reserve, though the details vary. During an actual outage, an enrolled battery stops supporting the grid and prioritizes your home. It's worth checking a program's minimum reserve level before signing up.

It depends on the program. Common models include bill credits, per-event or per-kilowatt-hour payments, upfront incentives on equipment and, in Australia, higher feed-in tariffs for exported power.

Big enough to rival conventional plants. Kraken's platform controls more than 5.7 gigawatts, Next Kraftwerke aggregates more than 10 gigawatts, and the Sunrun and Tesla dispatch in California delivered 580 megawatts from home batteries in a single event.

No. Most VPP resources are clean, like batteries, solar and flexible demand, but some VPPs also aggregate backup diesel or gas generators and bioenergy plants. What makes something a VPP is how it's coordinated, not what fuel it uses.

By Charles Kennedy for Oilprice.com

Related Posts

QatarEnergy eyes more US LNG deals to shield supply from Mideast conflict – report

US warns of ’ongoing cyber threat’ to industrial refrigeration systems

Factbox-Companies most exposed to potential US data-center curbs after blue sweep