We’ve reached a massive milestone in the global energy transition — the ability, right now, of renewable energy sources and energy storage to match or beat the price of power from natural gas-fired peaker plants and coal-fired generators.
The problem is that energy markets are not exactly free markets — and replacement of existing generation is not entirely tied to price. There are artificial regulatory constructs and financial structures built into energy markets which can favor certain parties (utilities) and fuels (legacy coal, gas and nuclear).
So, victory in the economic realm (increasingly the case with solar, solar-plus-storage and wind) is no guarantee of market victory if the regulations are stacked against renewables.
In addition to regulatory bias against renewables, one of the massive challenges of the energy transition is the need to equitably retire thousands of existing and planned fossil-fuel plants.
The fall of U.S. coal
About 85 GW of U.S. coal capacity has been shuttered since peak coal in 2007 — and the retirements have not slowed, even under a coal-sympathetic Trump regime. Actually, the pace seems to have quickened — with 31 GW of coal power retired since the beginning of 2018.
U.S. coal power generation fell by a dramatic 30% in the first six months of 2020, according to the U.S. Energy Information Administration (EIA).
While electricity from renewables grew 5% in the first half, the real beneficiary of coal’s fall was natural-gas generation which grew by 9% in the continental U.S. in the half, according to EIA. Natural gas gained a record share of the U.S. electricity energy mix in July of this year.
EIA said that coal plants are simply “uneconomical in most regions,” because of low-cost renewables, plummeting Henry Hub natural-gas spot prices and investors intently putting their money elsewhere. Nevertheless, about 236 GW of coal capacity is still active in the U.S. according to the Rocky Mountain Institute.
Outcompeting gas out West
The withering of the U.S. coal industry has an air of inevitability, whereas the existing (and planned) U.S. natural gas fleet will be a persistent part of the energy mix for decades.
Still, interconnection queue data, especially in the West, show new solar-plus-storage projects outcompeting gas plants. California’s interconnection queue included 29 GW of proposed solar-plus-storage plants as of year-end 2019, while Western utilities’ queues had another 33 GW. Yet planned gas units totaled only 0.2 GW in California’s queue and 4 GW in Western utilities’ queues. (data from Lawrence Berkeley National Laboratory report “Hybrid power plants: Status of installed and proposed projects”)
While the majority of plants proposed in interconnection queues are not subsequently built, a Western trend favoring solar-plus-storage over gas units is obvious. The Midwest shows signs of a similar trend, as the region’s interconnection queues (ERCOT, MISO and SPP) had 21 GW of proposed solar-plus-storage plants at year-end 2019, well above the 13 GW of proposed gas units. Gas is still favored in the East, as Eastern queues had 59 GW of proposed gas units at year-end 2019 and only 13 GW of proposed solar-plus-storage plants.
An additional finding: solar-plus-storage has 99.8% of the capacity value of a theoretical “perfect generator” in California’s CAISO grid region, according to a study commissioned by California’s three major utilities — easily matching gas.
Financially and socially equitable transition
Retiring the uneconomic coal units among the nation’s 236 GW of coal capacity (that’s most of them), and replacing them with solar, wind and storage, would yield savings of $10 billion per year, according to a report (How to Retire Early: Making Accelerated Coal Phaseout Feasible and Just from the Rocky Mountain Institute and Sierra Club.) Retiring all of them now, including units that are not yet uneconomic, would still save $9 billion per year.
But U.S. utilities cannot strand the assets of more than 1,000 peaker units and 236 GW of coal capacity without a different regulatory and financial structure.
The authors of the report recommend federal involvement in financing this transition, and using the savings not only to lower customers’ electric bills but to fund transition assistance and retraining for affected workers and communities. The authors recommend a federal program of loans or loan guarantees to refinance uneconomic coal plants, enabling plant owners to recoup their investment as they shift to renewables.
Some measure of systemic environmental racism can be reversed by replacing high-emission peaker plants, which tend to be located in areas with large proportions of minority and low-income residents, with energy storage.
Bias toward gas
“Many utilities are in a rush to acquire new natural gas-fired capacity, and clinging onto coal-fired generation” according to a report from policy shop, Energy Innovation.
After examining ten utility procurements, they found “utility preferences for gas-fueled generation” that “may be at odds with economics, but are not surprising.”
Utilities owned and operated about 1,900 gas units as of 2018, the authors found, and utilities’ preference for gas-fueled plants may be related to biases towards over-procurement of capacity and self-built generation, and “an organizational culture and rate design that favor gas-fueled generation.”
Also, “while utilities have generally acknowledged the value of grid services,” they said, “those values may not be recognized for new technologies in the same way that they are taken for granted from gas-fueled generation.”
Project examples
In May, utility Southern California Edison (SCE) signed seven contracts for a total of 770 MW of lithium-ion battery-based energy storage — to replace four large coastal once-through cooling plants. The award recipients are Southern Power, NextEra and TerraGen.
It’s one of the nation’s largest energy storage procurements and an indication of utility acceptance of massive-scale solar-plus-battery storage as a replacement for fossil fuel generation. Late last year, the California Public Utilities Commission (CPUC) urged California’s power providers and community choice aggregators to procure 3.3 GW of storage and PV-plus-storage systems to solve grid congestion and to compensate for gas and coal plant retirements.
Remarkably, SCE wants these energy storage resources online by August 2021, an aggressive timeline unthinkable for any type of fossil fuel project of this size.
Most of the winning storage projects are co-located with nearby solar power plants to charge the battery over the term of the contract, help integrate renewable energy into the grid, and furnish resource adequacy during peak demand.
An expansion of another enormous storage project in California was approved in August that could bring the Vistra Energy-backed battery for the PG&E-backed Elkhorn Battery Storage Facility in Moss Landing to a spectacularly sized 1,500 MW/6,000 MWh.
The battery modules are substituting for natural gas and with some poetic justice, are situated within an old power plant turbine building.
In an era of plunging battery and PV costs, these awards point to the dawn of dispatchable solar and wind as cheaper and cleaner than coal and gas.
Last word
Leah Stokes, a professor in the Department of Environmental Studies at UC Santa Barbara told pv magazine, “If you’re not building the future, you’re continuing to invest in old technology. And, that will have a cost at some point when you use billions to invest in coal plants, instead of building new wind, solar and storage that would create longer life value.”
***
Will Driscoll contributed to this article.
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: editors@pv-magazine.com.
These solar + storage assets are white elephants Eric. This is CSP + salt all over again. It didn’t work in 2008 and it’s not going to work in 2020.
The Moss landing project has two pumped storage projects within 100 miles. Those two existing storage projects are barely being used. Why are we installing another storage project in the same area?
You’ve been hard on Bloom over the years. These battery projects are every bit as bad if not worse because they’re pretending to be cheap. Six GWh is over a billion dollars. Where are the rate-payer advocates? They must be loosing their minds right now. These projects are expensive, unnecessary, distractions.
Here’s another reason 2020 is like 2008/2009 all over again. Member 2008. Member what happened to poly prices. Back in 2008 the economy nose-dived. This ruined a lot of peoples’ lives. There was a big push to get people back to work and as part of this push we prematurely built solar projects that shouldn’t have been built until 2012 or later. This resulted in billions of unnecessary spending. Billions schmillions… Who cares? Leaders need to lead and all that.
Installing batteries today is far worse than installing solar in 2009. Not just a little worse. WAY WORSE.
1. The power system doesn’t need them. Sure wind and solar are intermittent but we’re only at 10% wind and solar. The existing system can absorb wind and solar up to about 80 to 90% before the solar/wind curtailment rates spike up. Why install batteries today at $$$ to solve a problem that doesn’t exist when you can install batteries tomorrow at $ once you’ve verified the problem actually exists.
2. We can use Electric Vehicles to drive down battery costs. This would be a much better use of tax credit for several reason. First the credit would apply to a far higher percentage of the population. Over 80% of the population have cars whereas only around half of the population owns a home that’s suitable for solar + battery. Second, the batteries would off-set upwards of 50 to 60 metric tons of CO2 over the course of the vehicle’s life (not tome mention NOx, PM, CO and assorted HCs). The batteries we’re currently deploying are actually increasing emissions!!!
https://www.vox.com/energy-and-environment/2018/4/27/17283830/batteries-energy-storage-carbon-emissions
3. EVs compete with stand-alone batteries and batteries lose. See the graph in the Vox article. Notice how there are hundreds of times more EV batteries getting deployed than stand-alone batteries. All (or most) of those EVs are going to be trying to charge when electricity is cheap. We can safely predict this is going to be when wind and solar are plentiful. This means EVs will always be stealing storage’s lunch money.
This isn’t rocket science. We’re deploying junk to try to cover up the fact that wind and solar aren’t perfect. Who cares if they’re not perfect. I don’t. This ITC money going to batteries should be halted and ported over to EVs or something that deserves our support.
Here’s an excellent quote from the Vox article.
“Hittinger told me on the phone, and I more or less agree, that the right way to grow energy storage markets is to deploy the hell out of renewable energy and let the need for storage determine its growth. It is still entirely possible that, through whatever mix of transmission buildout, smart-grid improvements, and market reforms, we’ll end up needing less storage than we think. Market pull should determine where and when storage is deployed.”
Joe Joe
I initially was concerned at your post but I think you are largely right, except in a few constrained nodes the US doesn’t need batteries yet.
Although from work some of my colleagues in Australia are doing, some storage can still be very useful by just stopping peakers running at all or alternatively when used in a hybrid with wind and solar, they can allow the replacement of a baseload plant with wind solar batteries and a rarely used peaker. We did an example where we replaced a 1400 MW coal plant with wind/solar 500 MW gas and 600 MW batteries. Emissions were reduced about 95% and the batteries meant that most days the gas wasn’t even turned on. No batteries means the gas runs twice a day 250 days/y
However it appears that something like 10% of peak demand for two hours is all that is needed and there is little need for even that until VREs approach 50% market share
I think there are many ways stop peakers from running. This is another use case where batteries are being over-promoted. Here’s a good way to estimate the potential of eliminating peakers simply by installing solar.
1. Get the hourly data for some section of the grid in Australia. A large grid with good interconnection.
2. Put the data in an excel spreadsheet running vertically (solar, wind, coal, gas, hydro, other, total load).
3. Add a column called Residual load (Residual Load = Total Load – Wind – Solar.
4. Now Double the amount of solar and recalculate the Residual Load. Better yet scale up wind and solar until they supply 50% of annual load and then recalculate the Residual Load.
5. Now calculate the Peak Residual Load… Now compare the Peak Residual load to the historical Peak and estimate how much peaking capacity you need.
I’ve build many models that do this sort of calculation. What happens, as you probably know, is that the peak residual is much lower than the traditional system peak. The Residual Load peak moves to the 7 to 9 PM time period vs. the traditional system peak which reliably occurs between 4 and 6 PM. The reduction in the Residual Load peak is going to be the thing that puts batteries out of business.
One could argue… Well, what about using batteries to help reduce the residual peak? Meh… Sure… You could also use Dynamic Rates, Thermal Storage or improved building codes at a small fraction of the cost of batteries.
There was an interesting model NREL developed which can be used to estimate the A/C load on the power system. Basically what you do is take moderate load days and compare them to hot days and then you assign the extra load to A/C. You can take this hourly A/C load and put it in the spreadsheet I mentioned above with the hourly solar, wind, coal, load data. Then you can say… Hmmm… We’ve got X GW of cooling load in the Residual Peak.
Tesla Powerwalls cost $750/kWh installed… Thermal storage costs range from as little as 10 cents per kWh up to $10 per kWh.
So when I said here are three good reasons you said… OK… What now? But then when you read it you thought… OK… That actually makes sense?
“The reduction in the Residual Load peak is going to be the thing that puts batteries out of business.”
Pardon me… I meant to say the reduction in Residual Load is going to be the thing that puts peakers out of business. Maybe not all peakers but lots of peakers. This leads me to believe batteries are chasing a use case that is structurally shrinking.
The US Energy Information Agency started publishing hourly generation data for the lower 48 back in 2018. This data is broken down by region and resource type. You can use this data to guesstimate how much backup capacity a grid dominated by RE will need. Berkeley estimated the number at 360 GW but they also found that 70 GW of this total would be used less than 1% of the time. You obviously wouldn’t want to replace these 70 GW with batteries because that would cost a lot and you’d barely be using them. My bet is on dynamic pricing and flexible loads.
Final note… Overall the Berkeley study was well done study but they made several rather big errors. First off they had some strange projections for coal that are very out of date. Second off they didn’t include EV load. If you include EV load and then assume that most of this load will be following RE production as much as possible this makes it so your curtailment rates will be lower. But then if you follow the reasoning through this means we can build more RE because you’re not curtailing so much production. But then if you build more RE it means some of that 70 GW of rarely used gas (super-peak) will probably be replaced. I’d bet money we’re going to see a study out of MIT, Caltech, Oxford or Tsinghua which runs through multiple scenarios that have large amounts of controllable EV load and maybe even dynamic pricing. These types of studies will lay bare how out of touch a a lot of the grid battery projections are.
Eric pointed out an interesting technology to me a couple years back. Basically it’s a combined cycle plant with a thermal storage component. Basic idea is you charge up the thermal reservoir and then power the steam section of the plant with it – i.e. You don’t need to use it with the gas turbine portion. At the time I didn’t think the technology made sense but recently I’ve been thinking… Hmmm… Well if we need the gas capacity anyway (and I believe we do) then the thermal storage component might improve the economics of these plants because you’d be using them more. This is something that could be modeled out but I haven’t run the numbers yet.