The grid took center stage in Australia again last month, as a record heat wave and a coal plant failure threatened to topple an electrical system in the midst of a dramatic transition.
The grid didn't collapse, but high-alert events are becoming regular affairs for the large island nation of roughly 25 million people. Coal plants are shutting down, wind and solar are ramping up, and the shift from baseload to intermittent power has been punctuated by hugely unpopular statewide outages.
Energy storage could play the role of usher on the way to a new, cleaner grid. Or maybe it’s more of a firefighter, putting out conflagrations as they arise on that journey.
The latter better describes Tesla's Hornsdale project, the largest battery in the world, built in less than 100 days at the behest of South Australia's leadership.
Hornsdale put Australia on the map for big battery development, but it remains an anomaly in terms of both size and process. Since then, grid-scale batteries have become more routine, but it’s still hard to say a proper market exists; major projects thus far have required government support, rather than paying for themselves entirely with contracts or market revenues. That has now started to change, however.
Australia’s world-leading home solar penetration levels have spurred a healthy residential battery business too, as solar customers seek to maximize their investment as generous feed-in tariffs fade away. The commercial storage sector hasn’t seen similarly propulsive growth.
Conditions on the ground make storage more compelling for the Australian grid than almost anywhere else. With the right policies and savvy work by pioneer developers, Australia's energy storage market could not only become profitable, but also emerge as an example for the rest of the world.
Island in the sun
Australia’s grid is circumscribed by the coast. Australia is the sixth-largest country in the world, but that isn't much help when it comes to managing the grid.
Like the United Kingdom and South Korea, grid operators have to balance supply and demand without leaning hard on electrical imports and exports, as the U.S. and Germany are more free to do. It must do so while wind and solar power gradually displace coal, a process which introduces more intense swings of supply and demand.
The extreme environment makes this challenge more acute than it might be in a temperate zone. Australia gets hot (the recent heat wave topped 115 degrees Fahrenheit), which prompts spikes in demand for cooling.
Furthermore, the ruggedness of life in the interior has guided settlement almost entirely to the coasts. As a result, the national transmission grid doesn’t interweave freely across the landmass; it moves in an arc around the perimeter, with few and constrained points of connection between states that lead to flagrant disparities in electricity prices.
Rounding out the electrical landscape, Australia’s National Energy Market is energy-only, much like ERCOT in Texas. It doesn’t pay for capacity resources to be online in a pinch; the price swings, up to AUD $14,500 per megawatt-hour, are meant to incentivize private actors to invest in generation.
Wires utilities can’t own generation, but competitive retailers can, creating a class of firms dubbed "gentailers." The largest of that category are AGL, EnergyAustralia and Origin Energy.
Coal still rules the electricity mix, with 63 percent, followed by natural gas with 20 percent, according to the according to the government’s Australian Energy Update for 2018. Renewables deliver 16 percent, with hydro at 6 percent, wind at 5 percent, and solar at just over 3 percent.
Coal’s share has dropped in recent years; it delivered 73 percent as of 2014, but closures like Hazelwood brown coal plant in Victoria and Northern Power Station in South Australia have taken a toll.
For those who dare
The beauty of Australia’s market is that storage can participate without extensive regulatory modifications.
The first handful of projects did rely to some extent on federal or state support, but they proved out the business model, and utility-scale systems announced in recent months have penciled out economically on their own. Fluence announced in November that it will provide an unsubsidized 10-megawatt battery to a wind farm in South Australia.
The lack of capacity products means developers have to be comfortable with merchant risk and their ability to play in the market, which settles in 5-minute intervals. For those who dare, volatility offers its benefits. When prices surged toward the market cap, Fluence’s first Australian battery at Ballarat was ready, charged up on cheap off-peak power.
“It’s providing that peak capacity during that critical time when the prices are so high,” said Kiran Kumaraswamy, market applications director at Fluence.
Arbitrage can actually work for an Australian battery business case. When there aren’t massive price swings to take advantage of, batteries can play in eight varieties of Frequency Control Ancillary Services. Knowing how to optimize across these disparate revenue opportunities is the trick to successful operations.
“There are some very compelling returns to be had for a battery operating on a merchant basis,” said Matt Penfold, VP of commercialization at Advanced Microgrid Solutions.
That San Francisco startup originally focused on C&I development in the U.S., but has pivoted to algorithmic dispatch software for energy systems. It is finalizing its first deals in Australia, Penfold noted.
The key challenges, he added, are properly sizing a system to maximize return on investment, and gaming out how durable the returns will be in the face of new entrants jumping into the market and possible policy changes.
The first wave of batteries could prove very lucrative. By his calculations, a 20-megawatt, 1-hour system playing the markets in Queensland or South Australia could post a 20 percent internal rate of return. As in any market, margins will tighten as competition increases.
“Because it’s hard to understand, you’re going to see some more sophisticated investors chase that and make a lot of money," Penfold said. "Folks who have a little less risk appetite will put their money elsewhere.”
Additional roles arise alongside renewables plants. A wind or solar developer may also wish to control some of their output, making it dispatchable.
In an effort to protect grid stability, regulators recently decided that new renewable energy plants must be able to deliver 6 seconds of inertia for 47 percent of nameplate capacity, Kumaraswamy said. Every new plant will likely need a battery, but nobody builds 6-second batteries. The requirement could prompt developers to figure out what else they can do with this technology.
The Ballarat system showed that batteries could stack multiple uses for multiple customers.
Wires company AusNet owns it and uses it for reliability services, but also contracts its capacity to EnergyAustralia, one of the largest gentailers. The latter company serves its retail customers with a mix of power it generates and power it buys in the market. If prices spike, it can discharge the battery, offsetting some of the power it has to buy at elevated prices. The system acts as a hedge against wholesale price exposure.
On the horizon
In the more distant future, storage could take on a more robust peaking role, perhaps paired with wind and solar.
The lack of available capacity contracts often stands as a barrier to battery business models, like in Texas. However, capacity contracts also carry risk: What if a finite battery can’t discharge long enough to ensure local reliability?
The open market would allow battery peakers to make money as long as they have juice to bid in. It helps that natural gas in Australia is more expensive than in the U.S., which means batteries and solar could compete on cost there before they do in America.
Batteries could also play a more robust transmission role by easing the constraints between state markets.
South Australia leads in wind and solar development — it produced one-third of the nation’s wind power in 2017 — but it can’t export its full potential due to the constrained interconnectors to neighboring Victoria. That creates moments when prices are negative in South Australia but surging next door; this system inefficiency keeps costs higher and the power mix dirtier than it could be.
At a more technical level, Kumaraswamy explained, the interconnectors don’t operate at full capacity because an extreme rate of change of frequency (say from a major gust of wind appearing or disappearing) can damage the stability of the wires. However, he said, “If you put storage on the interconnector, you can easily manage the rate of change of frequency.”
The transmission use cases require additional policy work before they can become commercially viable. With the right clarifications, transmission used could drive hundreds of megawatts of deployments, Kumaraswamy estimated.
“Transmission processes need to value storage in their planning process,” he said.
Residential success spurs network approach
We’ve seen markets where big batteries rule (U.K., South Korea), and ones where small residential batteries are booming (pretty much just Germany, although the U.S. has really picked up in the last year). Australia uniquely offers a highly positive outlook for both home and utility-scale storage.
Commercial and industrial business models could work in theory, tackling demand charges, hedging direct buyers’ exposure to wholesale price swings or displacing diesel for off-grid mining concerns. That potential just hasn’t shown up yet in a big way.
“Residential and utility-scale storage are so attractive that C&I pales in comparison,” said Ravi Manghani, research director for energy storage at Wood Mackenzie Power & Renewables.
The recipe for residential storage success starts with Australia’s massive home solar penetration. Small-scale PV generated 10 times as much as large-scale PV in 2017.
But the generous feed-in tariffs that launched the distributed industry have cut down, and customers who got used to not paying a retailer for electricity have to grapple with their bills rising again. Batteries like Tesla’s Powerwall and sonnen’s Eco offer a way out.
I covered this market in detail for Squared subscribers, so I’ll direct you there rather than repeating myself.
The major residential developments since then revolve around the concept of decentralized networks of home batteries serving the grid, also known as virtual power plants.
In early 2018, South Australia awarded Tesla an improbably large contract to outfit 50,000 homes with solar and batteries over five years, totaling 250 megawatts of controllable capacity. That’s far bigger than anything Tesla has delivered to date, and the company has trouble balancing car production with timely Powerwall shipments in the U.S., so the announcement is best viewed with skepticism.
More credibly, major gentailer Origin Energy won a 650-home VPP deal in Victoria last fall, with state support. Rival gentailer AGL picked Enbala to provide network controls for a 1,000-home VPP. These programs offer retailers a way to win or maintain consumer interest amidst the competitive retail landscape.