A grid dominated by solar means power at night will cost 2-3 times day time rates. And demand for it will drop sharply.
That doesn't reduce the amount of nighttime generation capacity you need at all, unless people can shift demand from nighttime to daytime, which they largely can't. Most of the things that run at night either have to run 24 hours or have to run at night because you need light when it's dark and heat when it's cold.
It's a good augment their for electric companies to be the ones finance residential heat pumps and be able to count the installation as an asset.
> Adding higher nighttime electrical costs on top of that would not be helpful.
Higher nighttime costs would be definitely an issue with residential heat pumps.
I'm speaking from the UK where demand peaks are morning and evening and we don't have sun, so wind is the good match.
The problem is you're proposing a grid structure that I don't think is politically achievable. Solar is cheaper, so why not cover 100% during the day, wind is cheaper, that starts covering the night time, then suddenly your nuclear is uneconomic. And while you're spending the decades getting that power station built you're also having to explain why you aren't building out solar and wind to cover demand.
Because you still need something to cover the night, but if the thing that covers the night also generates during the day with minimal incremental cost, there is no cost benefit in unnecessarily duplicating that capacity again.
> wind is cheaper, that starts covering the night time
Except when it doesn't, and that's the problem. You have a huge wind farm that generates 50GW when the wind is blowing but only 5GW when it's still, and you have 25GW of minimum load at all times. On the night when you average 5GW generation for twelve hours, where does the other 240GWh of power come from? Your options are "build a peak 250GW wind farm instead so it never generates less than 25GW" (completely hopeless) and "build 240GWh of energy storage" (ouch) which then prices wind out because it's cheaper to use the batteries you need for a still night on every night and charge them with solar instead, and cheaper still to build 25GW in nuclear capacity.
> And while you're spending the decades getting that power station built you're also having to explain why you aren't building out solar and wind to cover demand
Because you are building them too. We can do two things at the same time.
the problem is solar is cheaper, so its profitable for someone to build 50MW of solar. so Nuclear covers night, now at twice the price. But wind is cheaper, so build 25MW of wind, maybe 50% of the time the wind doesn't blow, so nuclear only covers that, so it's now twice as expensive again. Now batteries are cheaper, so what's the role for that nuclear plant now? Would you invest in nuclear in that potential situation?
"Because you are building them too. We can do two things at the same time"
If it takes you 5 years to build the 25MW of solar, but the nuclear takes 15 years, theres a 10 year gap between solar build out finishing and nuclear being ready to take over from that coal plant. What do you do? 10 years of burning coal? or continue building solar so your covering as much demand as you can? But then why retire all that solar when the nuclear plant comes online?
That's not how that works. The unit cost of generation is inconsequential compared to the fixed cost, so you don't stop generating at any point and you just take the market price whatever it is.
Suppose nuclear needs to average $10/MWh to meet costs but solar can do $8/MWh during the day. Then as long as nuclear can get $12/MWh at night, it can match solar on price during the day and still cover costs. And solar + storage can't beat $12/MWh at night while the storage costs much more than $4/MWh. So then it's unprofitable to build the other 25MW in solar capacity because it would only get undercut on price during the day by nuclear which can make up the difference at night.
And it's the same thing for wind or anything else. Once you have 25GW of nuclear generation, nobody is ever going to be able to undercut it on marginal cost -- they'll never stop generating power and selling it for the market price, even if it would cause the plant operators to yield negative overall returns, because selling for an unprofitable market price is still less of a loss than getting nothing. And since everyone knows that ahead of time they don't bother to build a bunch of unprofitable wind turbines and that doesn't happen to begin with.
> If it takes you 5 years to build the 25MW of solar, but the nuclear takes 15 years, theres a 10 year gap between solar build out finishing and nuclear being ready to take over from that coal plant. What do you do? 10 years of burning coal? or continue building solar so your covering as much demand as you can? But then why retire all that solar when the nuclear plant comes online?
There is no hope of replacing the entire grid with solar in five years time. The production capacity for that doesn't exist, and would be unprofitable to create because once you've replaced the entire grid with solar all those new panel factories you just paid to build would go idle.
What actually happens is that you add solar as fast as you can and it's still not enough, and you add nuclear at the same time and if we're lucky then together they're enough.
Is it possible to minimize storage with nuclear? Yes, in theory, you spec the fixed generation amount for the peak demand and then have excess supply the rest of the time. Inefficient, but it works. This is clearly not the optimal mix however.
Is it possible to minimize storage with solar? No. It doesn't generate at night, that requires an enormous amount of energy storage to overcome, pure solar is a fail.
Is it possible to minimize storage with wind? Yes, but you have to match the peak demand with the minimum wind output. That's going to explode costs even worse than pure nuclear. Combining wind with storage to smooth out the load might be better, but by the time that's cost effective, so would using a smaller amount of storage to smooth the day/night load for nuclear and reduce the amount of generating capacity you need there too. Also, wind is not the cheapest generating technology to begin with (solar is cheaper).
So what happens if we combine them?
Solar plus wind is better, but still problematic. Wind generates at night, except when it doesn't. On a still night you could have to survive the whole night almost entirely on stored energy, but if you need almost as much storage capacity as for pure solar then you might as well make use of it every night and take advantage of the cheaper generation from solar than wind.
Wind + nuclear doesn't seem especially interesting either. The problem with wind in general is that its output varies randomly and doesn't track demand whatsoever, so you end up needing a large amount of storage proportional to the amount of wind generation you have. And wind isn't that cheap and storage is really expensive.
But solar + nuclear is excellent. The demand for electricity is higher during the day, which is when solar generates. Solar doesn't generate at night, but nuclear does. So you spec enough nuclear generation for the minimum nighttime load and use solar to both handle the higher daytime load and charge up the much smaller amount of energy storage you need to handle just the baseload-to-peak load differential in the few hours between when solar stops generating and people go to sleep, instead of the whole demand for the whole night.
You can also then price electricity higher during the night, which makes perfect sense when the solar that generates only during the day is the cheapest generation technology, and then nuclear pays for itself by beating solar+storage on cost for twelve hours a day while allowing you to cut your required solar generating capacity more than in half for the other twelve hours.