The distributed solar industry is diverging from “just solar.” A growing roster of companies talk about adding energy storage to solar and then aggregating those assets into distributed power plants. The appropriate business model for that concept, however, is very much up for debate.
Two camps have formed.
On one side: the commercial installers. They deal with bigger customers and bigger capacities, and seek out clients with fleets of buildings to maximize return on customer acquisition investment. For several years now, some companies have sold commercial solar, and a handful of others sold commercial storage; those two groups are converging at last.
On the other side: the residential players. They already put solar on people’s roofs; now they’re adding batteries and turning homes into miniature, dispatchable power plants.
Both camps have the same goal: to make customers happy while leveraging private assets as resources for the grid at large. Cleantech fortunes will be won or lost depending on which approach pays off.
The case for commercial
A handful of companies have been putting batteries in commercial properties for the last few years, including AMS, Stem and the company formerly known as Green Charge Networks. But commercial solar-plus-storage has picked up recently as costs and incentives hit the sweet spot.
Two developments of note: Stem, which had long dealt with solar for select customers, has made solar-battery pairings an official business priority, and SunPower is making batteries increasingly central to its commercial solar value proposition.
Roughly one-third of SunPower’s U.S. commercial solar proposals now include storage, said Rob Rogan, senior director for strategy and business development.
“In markets where you’ve got high demand charges or a big spread between peak and off-peak rates, a solar-only value proposition saves the customer some money, but adding storage can really turbo-charge the savings, to a point where they're not comparable with solar-only proposals,” he explained to me in an interview at Solar Power International.
“It's moving from optional in RFP requests, to now it's a solar and storage RFP request together. So if you don't have a storage solution, it's very hard to win those jobs,” he added.
SunPower will pilot aggregated grid services in 2019, working at first with aggregators that have existing contracts, Rogan said. The company delivers solar and storage to both residential and commercial customers, but he thinks the virtual power plant concept makes more sense at the commercial level.
“Having a portfolio definitely gives you an advantage, but it also needs to be a portfolio at scale,” he said. “To get to 20 megawatts of controllable storage, or a hundred megawatts, you can have a certain number of commercial installations, or an order of magnitude, or perhaps two, [more] residential installations to get to that same controllable capacity amount.”
LG Chem’s home battery and Tesla’s Powerwall both pack 5 kilowatts of discharge capacity. In contrast, Stem routinely installs 500 kilowatts to 1 megawatt at commercial sites and only occasionally dips down to around 50 kilowatts on the small end, said CTO Larsh Johnson. Based on those numbers, it would take 100 aggregated home batteries to equal the power capacity of one typical commercial battery.
Beyond the sheer size of C&I installations, they also develop from negotiations between the provider and a professional energy manager.
“They can understand the ins and outs of those offers in a relatively sophisticated way,” Rogan said. “And it's a fully dedicated asset: They don't expect it to do anything except save them money, and in some cases, deliver renewable energy credits.”
That distinguishes commercial aggregation from residential, where the homeowner typically buys storage for emergency backup power and has little understanding of what a virtual power plant even is.
It’s too early in development to know if this will be a problem, but it’s plausible that use of home batteries for grid services could conflict with their availability for the homeowners, either by degrading battery life or using energy before an unexpected blackout.
Compensation models for homeowners participating in grid services remain vague beyond the promise of lowering the upfront price in expectation of future revenues. Commercial energy managers, on the other hand, have the expertise to negotiate and codify revenue sharing that uses equipment installed at their sites.
If that company operates multiple buildings, the customer relationship could lead to many projects. Nearly 300 of Stem’s customers are enterprise accounts, Johnson noted.
The case for residential
There’s no denying the advantages of scale in an aggregated grid services fleet; fewer installations yield more capacity. But by that logic, society should build the biggest power plants possible and dispense with all this small-scale stuff.
Distributed generation has flourished, despite being more expensive per kilowatt-hour than large-scale central plants. Setting aside the merits of subsidies for small-scale solar, if private citizens choose to spend their own money in a way that reduces the need for communally purchased generation, ratepayers needn’t stand in the way.
We caught a glimpse of that future this summer, when Vermont utility Green Mountain Power called on its network of 550 Powerwalls in the homes of customers who opted to pay $15 a month for backup power. GMP used the aggregated capacity to shave its annual system peak, saving ratepayers $510,000 in an hour.
Elsewhere, utility Southern California Edison awarded Swell Energy a contract to deliver a 3,000-unit home storage fleet. Rooftop solar leader Sunrun has made storage a priority, and now packages it with all its solar deals in Hawaii and Puerto Rico, and half its deals in Southern California. The company is testing out aggregated grid services at a small scale in California and Massachusetts.
I called up Sunrun’s VP of Energy Services Audrey Lee and posed the aforementioned arguments to her.
“I’m happy to bust some myths here,” she said. “Residential has been fashionably late to the party, but we’re definitely part of the party.”
Just a few years ago, skeptics doubted that even large commercial batteries could provide grid services as reliably as front-of-the-meter assets, Lee said. She knows this because she pushed back on those doubts in her former role at AMS; now she’s extending that logic to home batteries at Sunrun.
“Residential assets are smaller, there are more of them, but the larger number of assets really isn’t a barrier on the software side,” Lee said. “Doing it for 100 or 10,000 is similar to me.”
It takes time to visit and install at all those homes, of course. But attaching batteries to the wall doesn’t take as long as getting a customer to sign up in the first place.
Furthermore, the professional energy managers that commercial vendors deal with can take months or a year to hammer out a custom contract suited to the needs of the business. Residential providers can use the same contract for all of their customers.
“You would think there would be more transactional cost with residential, but contracting with residential customers is easier,” Lee said.
As for the risk that Grandma won’t have power in a blackout because an aggregator used her battery for grid services, Lee said that’s the sort of technical problem her team is tasked with solving.
“We bake in the likelihood of an outage,” she said. “We don’t forget that we have obligations to the customer.”
The consequences of miscalculation are different in commercial aggregation, but still significant: A missed peak in that market could mean expensive demand charges. Commercial aggregators too have to balance grid revenue with protecting the interests of their host customers.
Why can’t we all get along?
Team Big and Team Small each has its arguments, and both deserve scrutiny if they are to become a new mainstay of the electric grid.
In practice, there’s no reason for grid planners to treat them as mutually exclusive.
“We’re not fighting for the same customers,” Lee said. “It’s a place where we should be cooperating to make something meaningful to the utility that’s less expensive than traditional poles and wires.”
Johnson, from commercial provider Stem, agreed.
“We are very focused on the space that we’re in,” he said. “If you look at what we’ll be able to accomplish in terms of scale of deployment…we’ll probably be able to get there faster than the residential markets will, but 'all of the above' is a good answer.”
If a utility needs an alternative to a pricey substation upgrade for a central business district, chances are there won’t be enough single-family households with solar to aggregate on that circuit. Commercial solar and storage could do the trick.
If the problem is load growth on a suburban feeder surrounded by houses, residential aggregation probably makes sense.
Programs covering a greater geographical space could well include both types of aggregated fleets. That hybrid model has yet to be defined, but it could involve utilities soliciting different classes of distributed energy resources and aggregating them into a portfolio, partnerships between different types of aggregators, or individual companies pulling together a range of assets.
E pluribus unum, but for the grid.