I remember there being a day earlier this year where the prices for wind-generated electricity in west Texas flipped. If you have to pay people to take your electricity (instead of spinning a generator down), why not store it for a later demand period?

The amount of energy that is stored in batteries like this isn't that big in grid-scale terms. For example, on the night that Texas wind generation flipped electricity prices into the negative back in September 2015, their wind turbines alone were pumping out almost 11,500 MW of power. This has a capacity of 129 MWh. It's useful for grid stabilization because it can start up almost instantly and fill in the gap until another generation source which can run for more than the 90 minutes the battery lasts can start up, but less practical for longer-term storage.

(The 90 minutes capacity at full output is pretty much an inherent property of lithium battery tech, as I understand it. There's a slight capacity vs power tradeoff which can be tweaked, but only within a relatively narrow band, and the maximum charging rate is if anything even less flexible than the discharge.)

Also, it looks like there's already a small-scale battery bank providing frequency stabilization using Samsung lithium ion batteries in West Texas: https://www.energy-storage.news/news/minimal-downtime-younic...

> The 90 minutes capacity at full output is pretty much an inherent property of lithium battery tech, as I understand it.

Even if that is true, that only means you need to expand your battery capacity and split into multiple banks.

It doesn't make sense to store something for later unless the storage costs are lower than the cost of reproducing what you need. A few periods per year of negative prices won't be enough to offset the investment in batteries.

I was under the impression that large scale energy storage is one of the big problems with renewable energy that has yet to be truly solved.

Don’t forget the stochastic control problem that needs to be solved to stop the whole grid blowing up!

Quick! Mine bitcoin for a day!

Superwog ep.5

This is immediately viable anywhere you're operating an inefficient single cycle peaker plant. The longer your battery can discharge for (4 hour, 6 hour, 8 hour, 10 hour), the more likely it's able to replace a peaker duty cycle when called upon.