A Shot of Storage

It got me thinking. The Energy Department’s expert on energy storage had just put up a slide, at NARUC’s Winter Policy Summit, saying that the nation’s grid decarbonization will need three hundred gigawatts of long-duration storage.

My mind instantly switched to calculator mode. Whoa. Three hundred gigawatts. It is a big number, sure. But the U.S. grid is gigantic. If it’s to generate mostly with wind and solar, are three hundred gigs nearly enough?

Do the math. Three hundred gigawatts of long-duration storage would deliver thirty thousand megawatts to the grid for ten hours. That is the minimum period, ten hours, that long-duration storage will produce power for the grid, according to Deputy Assistant Secretary Michael Pesin. Though he added last week at NARUC’s Summit, “… we also need to have storage that is over a hundred hours.”

Try this thought experiment. Suppose that three hundred gigawatts of long-duration storage were split up into three roles in the decarbonized grid of the future. In this scenario, a hundred gigawatts or a third of the total will be dedicated to delivering power to the grid for ten hours. Another hundred gigawatts, another third of the total, will be dedicated to delivering power for twenty-five hours. And the last hundred gigawatts, the last third of the total, will be dedicated to delivering power for a hundred hours.

This means that the first third will be able to deliver ten thousand megawatt-hours to the grid for ten hours. The second third will be able to deliver four thousand megawatt-hours for twenty-five hours. The last third will be able to deliver a thousand megawatt-hours for a hundred hours.

How do these capacities measure up to the magnitude of the problem? Is this enough long-duration storage to keep the U.S. grid reliably humming while wind and solar-based?

Consider PJM. It’s our largest control area. Though PJM is just a fraction of the entire U.S. grid.

During a mid-July 2021 week, PJM coal and natural gas-fired power plants generated around ninety-five thousand megawatt-hours per hour at daily peaks, while wind and solar generated around five thousand. Even at daily off-peaks, coal and gas plants generated around sixty thousand megawatt-hours per hour, while wind and solar generated well under five thousand.

No question we’ll be adding gobs of wind and solar to the PJM region in coming years. But no amount of renewable additions will be sufficient – after we take offline many tens of thousands of coal and gas generation – to carry PJM’s system through evenings and prolonged weather patterns adverse to wind and solar generation.

Deputy Assistant Secretary Pesin rightly said we’ll spread out the long-duration storage geographically. I’ve already cut the total into three slices for the three durations: ten hours, twenty-five hours, a hundred hours. Now we’ll cut each of these three slices into three slices for the eastern interconnection, western interconnection, and ERCOT.

Might the entire eastern interconnection end up with the ability to deliver something like four thousand megawatt-hours to all the east’s control areas for ten hours, two thousand megawatt-hours to all of the east for twenty-five hours, and four hundred megawatt-hours to all of the east for a hundred hours? Clearly, that’s not going to cut it.

The Energy Department’s Energy Earthshots Initiative for storage will detail its plans in a forthcoming report. We cannot wait to read how the report tackles the uber-important question of how much long-duration storage the decarbonized grid will need.

Unless we install a whole lot of long-duration storage, a lot more than the numbers above, it’s hard to see how the U.S. grid will perform reliably. Though that Energy Department report might show us how short-duration storage, perhaps, with its considerably greater capacities, can be exploited to complement and extend what long-duration storage can deliver.

But we may need to continue, for longer than many wish, to rely upon gas-fired plants to get us through evenings and prolonged weather patterns. Fortunately, the U.S. has an extensive infrastructure of gas plants, transmission, and storage.

There is a real possibility that green hydrogen will be ready in time to economically leverage this infrastructure. In our back pocket, fortunately as well, we do have a vast reserve of domestic gas together with the production capability to use it as necessary.