Can customers’ batteries, thermostats and EV chargers keep California’s grid up and running?

When it comes to building virtual power plants — groupings of buildings with electricity loads that can be turned down or energy resources that can be turned up to respond to the moment-by-moment needs of the power grid — multiple technologies can add up to more than the sum of their parts.

One example of this is the project that Clean Power Alliance, a community choice aggregator that serves about 3 million customers across 31 Southern California cities, is working on with VPP provider AutoGrid. It’s targeting 10,000 customers willing to host smart thermostats from Google Nest, ecobee and Emerson, electric vehicle chargers from ChargePoint and batteries from Sunnova that will be charged by rooftop solar.

Over the next two years, these partners hope to build up 6 megawatts of these resources that can be controlled in aggregate to ease grid emergencies like those that led to California’s rolling blackouts last summer. Outside of such emergencies, this aggregation could reduce reliance on polluting natural-gas-fired power plants to meet peak electricity demand and help customers save money on utility bills.

In those aspects, the Clean Power Alliance’s Power Response Program is like many of the VPPs popping up in California and around the country and the world. AutoGrid operates about 5,000 megawatts of such aggregations in 12 countries, using technologies from large-scale industrial loads in European energy markets to residential smart thermostats for U.S. utilities.

Any single technology can play a role in helping to balance the grid, AutoGrid CEO Amit Narayan said in an interview last week. But ​“at the portfolio level, you can squeeze more value out of what you are able to do with any one single asset,” he said.

Building the right mix of demand-side resources 

This kind of diversification isn’t a new concept. Demand-side management — a utility term for everything that happens on the customer side of the electric meter — has long employed a variety of interventions to reduce peak grid loads. These include energy-efficiency investments that reduce overall electricity use, demand-response technologies and programs that can cut power use during times of peak grid stress, and rates and incentives that reward or punish customers for using more or less energy at different times of the day or the year.

Similarly, utilities and independent demand-response companies have for decades relied on aggregated portfolios of individual commercial and industrial customers willing to cut energy use at times of grid stress in exchange for payment. Residential demand-response programs have also played a significant role in reducing demand peaks for many utilities, from old-fashioned radio-controlled switches to turn off air conditioners or pool pumps to more sophisticated programs that use smart thermostats.

But California has more solar-plus-battery systems and EVs than any other state. Last year, research firm Wood Mackenzie calculated that California has about 4.7 gigawatts of distributed energy capacity, about half of it from residential EV chargers, another third from customers signed up to reduce power use, and about one-tenth of it behind-the-meter batteries. The report projected this could grow to 13.5 GW of capacity by 2025, with most of that growth in the form of EV charging equipment and batteries.

A long-running research project from Lawrence Berkeley National Laboratory reported last year that with the right policies in place, demand-side resources could cost-effectively shift enough energy use away from the state’s post-solar evening peaks to replace the need for billions of dollars’ worth of batteries and other utility-scale resources.

Relying on millions of customer resources to replace power plants and battery farms is a risky prospect for the utility and grid planners accustomed to being able to deploy fossil-fueled power plants in a dependable way. Even so, California’s increasingly stressed grid conditions have put pressure on California regulators and utilities to find a way to tap these demand-side resources — and on third-party aggregators of those resources to prove they’re able to reliably perform the tasks the grid will ask of them.

Different technologies, different roles 

Smart thermostats are low-cost options for shaving a kilowatt or two of household energy use for an hour or two per day in a large number of homes, a fact that’s endeared them to utilities seeking mass-market options. Google Nest, for example, runs its Rush Hour Rewards program with Southern California Edison, San Diego Gas & Electric and the Los Angeles Department of Water and Power, and was able to reduce power use by about 60 megawatts during the state’s two-day grid emergency last summer.

Smart thermostats are also a good way to help lower-income customers save on energy bills. That’s driven utilities and third-party vendors including Google Nest and OhmConnect to offer steeply discounted — or even free — smart thermostats to customers in exchange for allowing them to respond to grid emergencies.

But smart thermostats are a bit of a blunt instrument when it comes to serving the grid. Raising a home’s temperature setting will reduce air-conditioning electricity demand — but just how much, and over what period of time, isn’t something thermostats measure, making their overall value a bit fuzzy to utilities and grid operators. And if customers decide it’s too hot, they can override those settings, which could leave the grid in the lurch without access to more directly controllable loads.

EV charging will be a far more common source of power use in the future. That makes it a critical target for utilities and other providers that must quickly reduce power use when the grid is under extreme stress. Enel X, one of the biggest demand-response providers in the state, also has a lot of EV chargers in its grid services portfolio. California’s $1 billion in utility incentives for EV charging infrastructure have regulator-approved requirements that the EV chargers that receive the incentives must be able to shift load when called upon.

At the same time, it’s hard to predict just how many EV drivers will be plugged in and willing to delay charging at any given moment and for how long.

Solar-charged batteries, meanwhile, are the most expensive resource in the mix. They tend to be owned by wealthier customers, making them a problematic target for direct utility incentives. Because battery owners rely on them to provide backup power during emergencies, programs that use them to bolster grid stability must be careful not to leave the batteries too depleted to meet that need.

But ​“there’s a firmness in the energy storage resource that you might not be able to find in the thermostat loads or the charging loads,” Michael Grasso, Sunnova’s chief marketing and growth officer, said in an interview last week. Solar-charged batteries often have plenty of spare capacity to provide during most of the year, he added.

This flexibility has made batteries a go-to resource for VPPs in California and other states. Many battery vendors are exchanging incentives of various kinds — in Sunnova’s case, a $1,500 gift card for new customers — for permission to tap their batteries to stabilize the grid.

Sunrun, the country’s leading residential solar vendor, has inked megawatt-scale deals in this vein with utility Southern California Edison. Sonnen, a solar-storage rival owned by Shell, has launched projects to install and aggregate batteries in multifamily housing in California.

Tesla is asking customers that own its batteries to allow them to be tapped during grid emergencies. And Swell Energy, a startup that aggregates other vendors’ batteries into virtual power plants, has lined up $450 million in financing to pull together VPPs in multiple states including California.

Fine-tuning incentives and market structures 

But what’s the best combination of third-party business models, utility programs and pricing mechanisms to optimize the value of these resources to help California’s grid? Utilities have traditionally been in control of demand-response and energy-efficiency programs. However, regulatory changes over the past decade or so have opened up third-party competition in states including California.

California’s Demand Response Auction Mechanism was initiated in 2015 to allow independent companies to create combined portfolios of behind-the-meter resources to earn money for grid services. While that program has been scaled back in recent years, growing numbers of residential aggregators have since taken the next step of participating directly in the energy markets of state grid operator CAISO.

Two of these providers, OhmConnect and Leap, have aggregated hundreds of megawatts of distributed resources to help balance the state’s grid.

AutoGrid has traditionally offered its software as a service for partners to manage VPPs — in fact, its software is being used by SunRun and Swell. But with its Clean Power Alliance project, it has taken the step of participating directly in CAISO’s energy markets, Narayan said.

Utilities ​“are somewhat constrained in how they bring all the different vendors and partners into demand-side flexibility,” he said. It can take years for utilities to win regulatory approval to launch new demand-response programs.

Community choice aggregators such as Clean Power Alliance have more flexibility, and they have used it. In the past two years, Northern California CCAs have launched programs aimed at enlisting megawatts of resources from solar-charged batteries, smart thermostats and other customer-sited resources.

“In recent years, we’ve witnessed California energy grids being overextended at unprecedented, alarming rates,” Jack Clark, CPA’s director of programs, said in a September press release. Its new virtual power plant program could be a ​“powerful tool in helping to serve and educate our customers, and ultimately reduce strain on the grid.”

At the same time, utilities are looking for ways to expand their role. Southern California Edison, the state’s largest electricity-only utility, has told state regulators it wants to transition to a ​“single demand-response program” that doesn’t force residential customers ​“to choose between competing [utility] programs” or opt for ​“one smart connected appliance or device over another.”

This, of course, could put the utility in competition with third-party aggregators that have long complained that existing regulatory structures have restricted their ability to maximize their grid value. Utilities and independent developers have sparred for decades over their respective roles in building power plants and renewable energy projects. Now they’re also sparring over the opportunity to tap the gigawatts of customer-sited resources that will be part of the energy future.