by Julian Spector
February 09, 2018

While solar-plus-storage is hot and getting hotter (just ask Xcel), you don't hear many savvy insiders talking breathlessly about the combination of wind power with storage. In fact, few people are talking about it at all.

The logic of intermittency suggests there should be a value in storing wind energy when it's plentiful and delivering it at times of relative need. That same logic has driven much of the excitement around solar-plus-storage.

The storage industry more broadly has evangelized its usefulness in a host of hybrid settings, including gas plants, hydro and seemingly any combination of renewables. The key is to identify the tangible benefits of putting a battery on the same site as a power plant, versus locating it somewhere else on the same grid.

In this week’s Storage Plus, then, we’ll examine why wind-plus-storage often gets left behind. Will it be consigned to obscurity, or will this market come from behind and surprise analysts the way storage has more broadly?

Wind isn’t sunlight

A core physical truth separates wind and solar power: predictability.

Some times of day, the sun is shining; other times, it never is. Rain and clouds interrupt the production pattern, but they don’t change the fundamental parameters.

Wind works differently, blowing at different times on different days and seasons. Offshore wind draws a steadier gust, but the point holds. There’s no one time of day when the wind starts and stops.

“You can have hours or days go by without wind, or periods where there’s so much you have to curtail,” said Daniel Finn-Foley, a GTM Research storage analyst who studies the hybrid market. 

One of the key business model advances for solar-plus-storage has been the ability to deliver a power-purchase agreement for clean power on demand, while benefiting from the 30 percent federal Investment Tax Credit. Two main reasons make it hard to carry that model over to wind.

“If you’re guaranteeing power for peak periods, wind doesn't work,” Finn-Foley said. You can’t count on wind to charge up the battery conveniently prior to the evening ramp.

The ITC gets complicated too, because wind farms also qualify for the federal Production Tax Credit, which almost always proves more lucrative. They’re not allowed to take both credits, so the PTC wins out.

“They make more money in the long term as opposed to offsetting a small amount of upfront costs,” Finn-Foley noted.

For the ITC to provide a better economic return than the PTC, the battery would need to be 4-hour duration and have higher nameplate capacity than the wind farm itself, according to modeling Finn-Foley did with wind analyst Hong Durandal of MAKE. Such an oversized battery system exists precisely nowhere in the real world.

There’s an argument to be made that wind power stored in batteries and then discharged to the grid should receive the credit just like power direct from the turbines. The industry hasn’t tested this premise, so there is little precedent to point to if the Internal Revenue Service asks for an explanation (proposals to create a standalone storage tax credit could clarify the legal ambiguities, but those haven’t gained much traction).

In short, the tax credit rules deny a key benefit to siting storage alongside wind. 

But what about arbitrage?

When I talk to wind developers about storage, I typically hear a rapid-fire dismissal that the arbitrage play just doesn’t make sense.

It’s important to note that batteries can do a lot more than just buy cheap power and sell it at high prices. But the doubtful assessment of this market opportunity does have data to back it up.

The researchers analyzed a 1-megawatt/1-megawatt-hour battery performing load-shifting in Texas at different price points. Today’s cost, including upfront capital, operations and maintenance, and replacement costs, sits around $1,235/kilowatt.

Such a system would lose millions of dollars over its lifetime. The good news doesn’t appear until all-in system cost drops to $335/kilowatt, expected around 2028. Even then, the system just squeaks by with a sliver of profit, and only in ERCOT’s Load Zone South.

By then the ITC will be long gone, so that benefit of co-location won’t even exist.

Let’s say you want to scale it up to two hours of load-shifting. Those numbers look even worse: The price goes up for the longer duration, but profits are nowhere to be seen. The price spread between ERCOT’s peaks and troughs is not sufficient to support a profitable wind-battery hybrid.

But what about ancillary services?

In storage developer parlance, ancillary services occupy a hallowed spot on the value stack, shining a beacon of hope for battery project economics.

They just don’t pay well.

In most markets, frequency regulation operates through a single signal, which tells a storage device to charge or discharge based on what’s needed. If the battery charges from the grid, it forfeits the ITC, which requires renewable charging.

A battery that charges primarily from a wind farm could claim the ITC, but then it can’t serve the Regulation Down market, which requires charging from the grid. Removing the ITC strips a major reason to co-locate.

ERCOT and CAISO split their signals into Regulation Up and Regulation Down, which could allow wind-coupled batteries to compete solely for Reg Up.

The clearing price for Reg Up service in ERCOT has hovered around $8/megawatt-hour for the last couple years, limiting the profit that a battery system could hope for. Risk of oversaturation of the market also limits potential returns (just look at PJM).

But forgoing Reg Down revenue, in Finn-Foley’s analysis, keeps a battery’s internal rate of return down below 6 percent. It’s more lucrative to reject the ITC and just have a grid-tied battery, but then there’s no need to put it on a wind farm.

Where does it actually make sense?

As we have seen, several of the marquee energy storage roles don’t fit well for wind-farm-located batteries. Other jobs can just as easily be performed elsewhere on the grid, meaning co-location fails to provide an added benefit.

“There isn’t really a business model for wind-plus-storage, and I don’t see one developing in the next five years,” Finn-Foley said.

There are a few specific cases where the combo makes good sense now or in the near future, and a few that will come into focus as years pass and the grid changes.

Site-specific ramp-control issues: On islands with bountiful renewables and nowhere to export to, a battery on a wind farm can do a lot of good. It can smooth the ramp-up as wind picks up, and jump in if the wind suddenly goes slack. As islands adopt more wind, this could become a more regular grid-balancing role. In most places, though, a market product for that service does not yet exist.

Ease of installation: If a developer wants to deploy a grid-scale battery, and already has a wind farm with a good interconnection, it could save time and money by sticking the battery there. There might be savings from labor costs by combining projects as well. This is more of a balance-of-system play than an operational play. Look for this from big renewables firms like RES, NextEra, E.On or Invenergy.

Microgrid with lot of wind: Having wind fluctuations in a small, contained system would make batteries more useful (similar to the island case). But good luck finding a wind-heavy microgrid; predictable energy output is the whole point, so they tend to rely on solar, diesel gensets or combined heat and power.

More wind on the grid: Batteries paired with wind will become more valuable as the share of wind power on the grid increases. Until that happens, Finn-Foley said, battery storage for wind is solving a problem that doesn’t really exist yet.

“Right now, wind energy made in the U.S. is pretty much being used,” he noted. “Even in Texas, there’s not enough curtailment to make storage viable.”

If there’s significant wind resource pouring onto the grid at once, it will bring curtailments and lost profits for developers, especially for new projects in areas already struggling with curtailment. That shifts the cost-benefit analysis for storage.

More volatile wholesale markets: Let’s fast-forward 20 years to a Texas with significant wind saturation. This change will spur longer periods of negative pricing when a major gust gets all the turbines cranking, as well as more severe ramping needs when the gust vanishes. That’s a much tastier recipe for arbitrage.