Hacker News new | ask | show | jobs
by pyrale 1615 days ago
> But nuclear power is also not really suited for load following.

That's a common assertion on that topic, but it is baseless. In fact, nuclear plants are routinely used for load following when needed [1].

[1]: https://www.oecd-nea.org/nea-news/2011/29-2/nea-news-29-2-lo...

1 comments

I think the issue with load following is that as most of the costs of nuclear are fixed costs, it increases the cost per unit of energy in proportion to the amount you "turn down" the output. As nuclear is already struggling to be cost competitive with solar+wind + storage, that makes it hard to justify.
Cost per unit of energy is talked about a lot, but it's not that relevant, though. What people buy is usually guarantees to be able to provide. There are different types of contract (peak power vs base load, futures vs. forwards, spot vs. month- or year-ahead contracts), but overall, what people buy is a guarantee that they will get power when they need it. Whether they actually use it or not is not as relevant from the contract's pov.

The ability to load-follow is unrelated to energy price, but to grid stability. Regardless of what contract exists between a provider and a consumer, there is a third party, the grid operator, who can ask the generating parties to adapt their production to actual consumption.

> solar+wind + storage

Simply doesn't exist at scale currently. We don't really have good estimates of what a storage-balanced grid costs at scale, and we don't have the industrial bandwidth to build storage at scale with the current technologies.

To give you a back-of-the-envelope calculation, current estimates are that european countries relying on wind/solar would need 8 days worth of batteries to avoid most of negative-generation events. For a country like Germany which consumes 1.5TWh a day on average, that would be more than many thousand units of the large battery Tesla built in Australia.

The scale of this storage is tied to variability ('intermittency'), and at least a way to reduce it is known: spread out the production units.

Case in point: wind in Europe https://www.imperial.ac.uk/news/180592/european-cooperation-...

This would lead to marginal improvements, but wouldn't change the fundamental scaling issues with renewables and storage.
This is an opinion.

"the planned development around the North Sea means 100 GW (100 large power stations) would need to be turned on or off to balance out changes in wind power production when the weather changes. With a more cooperatively designed system, this could be reduced to just 20 GW across the continent.". Add solar, biomass... and storage, including a smartgrid enabling (for example) V2G: https://en.wikipedia.org/wiki/Vehicle-to-grid , then green hydrogen (boosting production units' output and reducing the amount of electric electricity needed).

8 days' worth of green hydrogen to feed gas turbines would fit in a few relatively small gas storage facilities.
We don't have production facilities for that much clean hydrogen either (about 350.000 tons would be needed for 8 days' worth of Germany's needs. World production of green hydrogen was about 1 million tons between 2015 and 2018[1]).

As I said earlier, the theory may exist, but solving scaling problems isn't trivial, and is going to take some time.

[1]: https://www.statista.com/statistics/859104/hydrogen-producti...