Image: Heisenberg MediaEveryone already knows the big news this week was that Tesla Motors will buy SolarCity Corp. for $2.6 billion -- a move that helps Tesla improve its economies of scale when it comes to manufacturing battery storage for both cars and energy. It's a move that, generally, has positive implications for the development of battery storage across the board, although this story will focus on storage used for power generation.
“Together with Solar City, Tesla will bring more electricity from renewables direct to consumers and onto the grid," says Jostein Eikeland, chief executive of the Alevo Group in Charlotte, NC. And, he added, distributed resources is a good thing for the environment as well.
Residential devices are smaller systems that produce 5-10 kilowatts of electricity and often installed in garages. More expensive per kilowatt hours, they are typically used for backup power (instead of generators) or to maximize rooftop solar production. Grid-scale storage is 1 megawatt or more and sold to utilities by storage developers, typically through requests for proposals. They are used for frequency regulation or renewable smoothing – the most prevalent use today.
As for Tesla, it will sell its residential storage batteries for about $3,500, although the cost does not include “inverters” that manage voltage levels during charging and discharging. They are lithium-ion batteries, which can discharge for an hour.
At least 40 different storage technologies already exist – but utilities’ demands are stringent. They’re trying to match the energy they produce with the energy their customers consume. That’s hard to do, given that industrial sites routinely implement new processes.
Utilities often have to overbuild so they have capacity when demand peaks. If more storage could be developed, the grid could prevent blackouts with less excess capacity, keep wholesale prices of electricity from fluctuating wildly, and entice more green energy projects.
"We have a number of storage assets across our system," says Steve Young, chief financial officer of Duke Energy, in a phone interview just after the company announced its 2016 second quarter earnings Thursday. "It is not a tremendous dollar investment. These are pilot projects and these storage technologies have a ways to go. They are built around renewables in the western United States. They are not cost effective at this time."
The goal, though, is to increase size and scale. And if with every iteration of development, the technologies are getting better and cheaper.
Beside Tesla, other battery companies are making a go of it in the utility business. With 130 employees, Pittsburgh-based Aquion Energy has started commercial production of batteries that allow utilities to better manage their grids. It says it is selling its technology to California and Florida utilities that it cannot name because of confidentiality agreements.
“We are storing energy from solar panels and integrating renewables onto the grid,” says Ted Wiley, Aquion’s vice president of products and corporate strategy, in an interview.
Aquion is providing the battery technology, for example, at one of Duke Energy's sites. That utility is now testing a first-of-its-kind battery technology at a substation in North Carolina -- one that is demonstrating rapid response, real-time smoothing and load shifting.
One of the distribution lines at Duke's substation has a 1.2-megawatt solar installation connected a mile away. With North Carolina fourth in the nation for installed solar power, the company says managing and maintaining these grid-connected renewable installations is critical now and in the future.
"This approach will allow our energy storage systems to do a variety of tasks," said Thomas Golden, technology development manager for Duke Energy, in a statement. "With so many solar installations in North Carolina, we must look for innovative ways to better incorporate renewable energy into our system – and still provide reliable service at a competitive price for our customers."
Altogether, the Electric Power Research Institute says that installed battery capacity is now at 887 megawatts, which is up from 300 megawatts three years ago. But that figure is expected to grow now that California is requiring its incumbent utilities to provide 1,325 megawatts of energy storage capacity by 2020.
Ideally, battery storage would allow wind and solar developers to grow their markets, make grids more efficient, and eliminate the need to expand power production to cover peak loads. But it’s expensive. Until the price of batteries comes down dramatically, progress will be slow.
Some green energy advocates say the only problem is scale. Once advanced batteries are produced in sufficient quantities, they argue, the cost of manufacturing them will fall. Tesla chief executive Elon Musk, for example, has said that his $5 billion “gigafactory” could bring down the costs of advanced batteries by more than 30 percent. His success would thus mean success for many others in the industry.
Consider Duke's “Notrees Energy Storage Project:” It is not what the average person might envision — a battery that soaks up the grid’s electrons at night and then releases that electricity during the day if the wind is not blowing. While that’s the long term objective, the battery’s main function right now is to feed the electrical network with the energy it would need if a power surge occurs that would cause the grid to overload and the lights to go out.
Simply, each time an industrial outlet or a residential customer uses electricity, the system operator is tasked with matching the demand with the existing supplies. And when something tips that balance, the “frequency” can fluctuate and machines will automatically shut down. That’s how cascading blackouts are started.
When the network gets out of kilter, the lead acid batteries used by Duke can almost instantaneously respond. “Storage is good at being fast and accurate at small deviations in frequency,” says Jeff Gates, with Duke's Notrees, who adds that the cost to both business and residential consumers is less than if generators were used to fill that void.
