> You can absolutely build solar+wind+storage systems that deliver 24/7/365 energy with many nines SLAs, on the real earth with real statistics on weather.
Once you start adding nines, nuclear starts to be become attractive again. Hence these deals.
Once you start adding nines, nuclear starts to be become attractive again. Hence these deals.
Only if you ignore reliability issues with nuclear, such as France having to take a lot of reactors offline at the same time to check for cracks. The comparison has to be fair.
Second, the could have kept those plants operating and taken a more piecemeal approach to fixing the cracks. They shouldn't have, and didn't, because you really want that triple or quadruple redundancy in your cooling systems, but they could have if push had come to shove, the plants were still operational.
But since there was plenty of capacity available, they could afford to take those plants offline and do the checks and repairs all at once.
Intermittent renewables afford you no such optionality.
If the sun don't shine, the sun don't shine.
If the wind don't blow, the wind don't blow.
These cracks are not the only factor for the poor performance on nuclear in 2022 : several reactors had to do their 10-year maintenance (grand carénage), which had been delayed because of the pandemic.
It's hard to extrapolate reliability issues from an event that happened once in 30 years.
Nuclear and reliable renewables for base load, intermittent renewables for intermittent load where and when it matches.
Wind/solar are not cheaper. Per kWh they are more expensive. Thanks to China, they are a lower and smaller-grained capital investment.
German und US plants had/have capacity factors of well over 90%, and the downtimes are scheduled.
The "overbuild" is minimal.
Dayjob is in energy market algo trading and DER ancillary services in EU, and I can tell you with confidence this is far from true.
And when the nukes trip for safety reasons - which happens multiple times per year - that’s a GW that just dropped off the grid from one second to the next.
For me, I much prefer the reliability of wind and solar; never seen a correlated failure, and we know their production pattern days in advance.
How these things will be managed in the real world matters, too, not just how it could play out hypothetically.
Until March 2023, there was a law in France that made it illegal to increase nuclear capacity in absolute terms and mandated reduction to under 50% in relative terms.
https://www.youtube.com/watch?app=desktop&v=isgu-VrD0oM
But again, a one-time event, and it doesn't change the high capacity factors world-wide (over 80%) and the extremely high capacity factors in reasonably well run fleets ( > 90% in the US, and also inGermany when they still had nuclear ).
2. A one time event has little to no effect on overall capacity factors.
3. Underinvestment is very easy to deal with: invest adequately.
Since we didn't have "no electricity for some months" in this one-off, and, as I explained elsewhere, had optionality, not quite sure how you extrapolate from an event not happening to an event happening every few decades.
Gosh, so we're not ever going to invest sufficiently in battery tech for that to ever work. Glad you cleared that up for us.
But to be clear, it wasn't some random "human nature" that caused the underinvestment. It was official government policy, and that government policy has now been changed.
The renewables mafia that is still in control in Germany was also in control in France. That has been taken care of, mostly by being exposed to the harsh light of Reality™.
The law that mandated reduction and forbade expansion of nuclear was repealed March 2023. They are now expanding capacity, build first 6 and then an additional 8 simplified EPR2 reactors.
And there was one more reason for the lull in investment: France did the initial expansion under the Messmer Plan far too quickly. They nuclearized their grid in just 15 years.
Which sounds great, and of course completely puts the lie to those who claim that nuclear is "too slow". But there is a problem: these nuclear plants are not just quick to build, they also last pretty much forever, at least 60 years, many if not most are being extended to 80 years and the experts say there's no reason not to go to 100.
If you have built all the nuclear power plants you need in 15 years and they last 100 years, you have nothing to build for 85 years (or 65 / 45 years if you assume 80/60 year lifespans).
That's a problem, because your nuclear industry will wither and die in that time.
Which it pretty much did.
Poor policy can happen again - such is human nature. So things need to work under poor policies, too.
Exactly, as I wrote. Which is why they were able pursue the option "shut everything down at once and fix immediately" without any negative repercussions apart from EDFs bottom line for that one year.
(Though that apparently was not the biggest negative impact on their bottom line that year. The bigger one was being compelled by law to sell electricity to their competitors at 4 cents/kWh and having to buy it back at market rates that were spiking due to the energy crisis).
There was also the option "stagger the shutdowns and take a slightly elevated risk of one of the triple-redundant cooling systems failing". Which they might have preferred to reducing electricity supply for the country or accelerating repairs. But they didn't have to do that.
> So things need to work under poor policies, too.
Nothing works under arbitrarily bad policy, and such a standard is not applied to anything in the real world.