Operating costs for nuclear power plants are low and consistent over time. The bulk of the cost is the initial investment, and of that the bulk is financing, i.e. interest.
This has a graph of operating costs for French nuclear plants over time on page 10. After an initial high cost of 40 centimes / kWh it settles down to slightly above 10 centimes / kWh in around 1984 and then stays flat until 2000, where the graph ends. Let's call that 12 centimes. The French franc was converted to € at a rate of 6.55 : 1 so that's slightly less than 2 euro-cents per kWh.
2. Capacity factors
Current French nuclear plants are old designs and there was significant overbuild. Modern plants easily hit > 90% capacity factors, heck, the EPR even has 4 independent cooling systems so that they can keep the plant running while doing maintenance on the cooling system! Alas, that's one of the reasons it is so difficult to build.
For the US: "Nuclear has the highest capacity factor of any other energy source—producing reliable, carbon-free power more than 92% of the time in 2021."
France is also adding renewables so the nuclear plants don't have to buffer all the variability in demand. Right now they have nuclear plants that they just shut down on weekends.
So let's low-ball this and say the capacity factor of these new plants will only be 80%. How much does this affect the calculations?
1.5 * 24 * 365 * 0.8 * 80 = 840960
So instead of producing 946 TWh over its lifetime, the plant will produce 840 TWh or 840 billion kWh. That doesn't really affect the calculations much. At 1 cent / kWh that's € 8.4 billion, which is still pretty close to the construction costs. At 2 cents / kWh we are looking at € 16.8 billion. Add another 2 cents for operating costs and we are at a total of 4 cents / kWh.
4 cents / kWh.
Now this is obviously not a perfect calculation, if such a thing can even be made. But it is very much in the right ballpark.
The claims of 20 cents / kWh or more by anti-nuclear lobbyists are not in the right ballpark. And unsurprisingly, their source for those kinds of numbers is "trust me, bro". Or Lazard, which is arguably worse than "trust me, bro".
3. Lifetime
Yes, I believe 80 years is extremely realistic, particularly for these newer reactors that were specifically designed for long operation, with lessons learned from older designs. Probably more of a lower bound than an upper bound.
"There are no technical barriers to running some nuclear plants for up to 80 years, ..."
"There are two research programs addressing the five main challenges to long-term operation: primary system metals and piping; concrete and containment structures; electrical cables; reactor pressure vessel and buried piping. “These programs have not uncovered any technical show-stoppers that would prevent the renewal of licenses from 60 to 80 years,” the study authors wrote, adding that more research is needed."
"... a majority of executives say that it is very likely their plants will operate for 80 years or longer. It is a fairly natural progression, according to Was.
"If they last till 60, maybe they can last to 80," Was said. "Heck, maybe 100?"
> Operating costs for nuclear power plants are low and consistent over time.
CRE‘s current cost estimates for operating existing nuclear reactors (Ie zero profit, and excluding all costs associated with construction, risk, and decommissioning) is ~€57.8/MWh. Some of that is age, but even with an already paid off power plant, and zero money set aside for decommissioning, and the government talking 100% of the risks of a major accident, and no need for profit, it’s still 3x the current total cost of solar per kWh.
Further that’s their optimistic estimate. I get you want to crunch some numbers here, but they just aren’t matching up.
> 2. Capacity factors
> Modern plants easily hit > 90% capacity factors
In other countries sure, in France reactors are more limited by Curtailment than just technical issues.
If nobody wants to use electricity from your reactor you must shut it down or damage the electric grid. If France had 32 GW of Nuclear they could have ~90% capacity factors but they have 62 GW and nobody wants that much power on nights and weekends. https://en.wikipedia.org/wiki/Curtailment_(electricity)
Now if you’re projecting France to have far less nuclear in 2070 then at first 90% seems reasonable. Except ultra cheap wind and solar is what’s going to replace nuclear and that makes things worse for nuclear with regular oversupply at zero marginal costs vs nuclear non zero fuel costs.
CRE is also having other issues, but I don’t think we can know there will be equal levels of mismanagement and incompetence in 70 years. So I’m mostly ignoring their recent issues.
> 3. Lifetime
Again not purely a technical issue. Power plants cost more to maintain over time and increased maintenance means lower capacity factors. You still pay ~€10/MWh for enriched uranium, and you still need a huge workforce, but most pumps etc last 10+ years, they don’t last 80. And you can’t replace a critical pump during normal operations so you end up with more downtime. Eventually concrete and steal will fail and there’s no cheap way to replace a steel pipe embedded in several feet of concrete.
Thus the 40-50 year lifespan, roughly when replacement + decomisioning costs less than maintenance. Ideally you line things need replacing on different years so you can fix em during refueling, and then have a bunch of equipment need replacing right after you shut things down. However France extended the plants lifespans so in effect they were hit with a maintenance backlog.
> own cost estimates for operating existing nuclear reactors (Ie zero profit, and excluding all costs associated with construction, risk, and decommissioning) is ~€57.8/MWh.
You are 100% incorrect. This is "the complete production cost of France’s existing nuclear fleet". And yes, I put that in quotes because it is a direct quote from the article you cite to claim the opposite. It's in the first sentence, difficult to miss. Furthermore, this is an estimate for the future total cost between 2026-2040.
And 5.7 Cents / kWh total cost for dispatchable, reliable and CO₂ free energy is very good.
Which is why France is investing heavily in nuclear again.
> Curtailment
"Nuclear power in France has a total capacity factor of around 77%, which is low compared to nuclear power plants in other countries due to load following."
So 77%, not the 50% you wrote. And they are increasing the share of other electricity sources therefore lowering the total share of nuclear (which was overbuilt during the time of the Messmer Plan). So curtailment of nuclear will decrease and capacity factors can only go up from the 77% they have.
Unless you give priority to variable renewables. Which you can do, but it's a stupid idea.
Given that, 80% total seems entirely realistic for France, and for the new plants they can go higher than that. Curtail the older plants that are already paid off. However, the key point is that this is a choice they can make. With variable renewables, you get average capacity factors in the 20-30% range, if you're lucky, and you do not get to choose.
Anyway, even at the completely unrealistic 50% you assume, the cost would be around 6-7 cents per kWh, which still seems fine given retail electricity prices of around 25 cents / kWh in France and above 35 cent / kWh in Germany. And of course it is nowhere near the 20+ cents anti-nuclear lobbyists claim with no evidence whatsoever.
> Thus the 40-50 year lifespan, roughly when a replacement costs less than maintenance.
Sorry, the actual experts on this + the owners of the plants disagree with your off-the-cuff speculation, and they are extending the lifetime even of their existing plants to 80 years and possibly beyond.
Operating costs for nuclear power plants are low and consistent over time. The bulk of the cost is the initial investment, and of that the bulk is financing, i.e. interest.
https://endexiresearch.wordpress.com/wp-content/uploads/2020...
This has a graph of operating costs for French nuclear plants over time on page 10. After an initial high cost of 40 centimes / kWh it settles down to slightly above 10 centimes / kWh in around 1984 and then stays flat until 2000, where the graph ends. Let's call that 12 centimes. The French franc was converted to € at a rate of 6.55 : 1 so that's slightly less than 2 euro-cents per kWh.
2. Capacity factors
Current French nuclear plants are old designs and there was significant overbuild. Modern plants easily hit > 90% capacity factors, heck, the EPR even has 4 independent cooling systems so that they can keep the plant running while doing maintenance on the cooling system! Alas, that's one of the reasons it is so difficult to build.
For the US: "Nuclear has the highest capacity factor of any other energy source—producing reliable, carbon-free power more than 92% of the time in 2021."
https://www.energy.gov/ne/articles/what-generation-capacity
In Germany, it was 93% IIRC before it went to 0% due to government action.
"In 2019, the global median capacity factor was 85.9 %, in line with the load factor in recent years."
https://www.iaea.org/newscenter/news/iaea-releases-2019-data...
France is also adding renewables so the nuclear plants don't have to buffer all the variability in demand. Right now they have nuclear plants that they just shut down on weekends.
So let's low-ball this and say the capacity factor of these new plants will only be 80%. How much does this affect the calculations?
So instead of producing 946 TWh over its lifetime, the plant will produce 840 TWh or 840 billion kWh. That doesn't really affect the calculations much. At 1 cent / kWh that's € 8.4 billion, which is still pretty close to the construction costs. At 2 cents / kWh we are looking at € 16.8 billion. Add another 2 cents for operating costs and we are at a total of 4 cents / kWh.4 cents / kWh.
Now this is obviously not a perfect calculation, if such a thing can even be made. But it is very much in the right ballpark.
The claims of 20 cents / kWh or more by anti-nuclear lobbyists are not in the right ballpark. And unsurprisingly, their source for those kinds of numbers is "trust me, bro". Or Lazard, which is arguably worse than "trust me, bro".
3. Lifetime
Yes, I believe 80 years is extremely realistic, particularly for these newer reactors that were specifically designed for long operation, with lessons learned from older designs. Probably more of a lower bound than an upper bound.
"There are no technical barriers to running some nuclear plants for up to 80 years, ..."
"There are two research programs addressing the five main challenges to long-term operation: primary system metals and piping; concrete and containment structures; electrical cables; reactor pressure vessel and buried piping. “These programs have not uncovered any technical show-stoppers that would prevent the renewal of licenses from 60 to 80 years,” the study authors wrote, adding that more research is needed."
https://spectrum.ieee.org/aps-argues-to-extend-lifespan-of-n...
"Research demonstrates no technical limits for reactors operating up to 80 years."
https://www.energy.gov/ne/articles/whats-lifespan-nuclear-re...
"A handful of reactors have already been approved to operate for a total of 80 years,..."
https://www.technologyreview.com/2024/04/04/1090630/old-nucl...
"... a majority of executives say that it is very likely their plants will operate for 80 years or longer. It is a fairly natural progression, according to Was.
"If they last till 60, maybe they can last to 80," Was said. "Heck, maybe 100?"
https://www.scientificamerican.com/article/nuclear-power-pla...
As an example, the AP-1000, which is a peer to the EPR/EPR2 designs is specifically touted as having an 80 year lifetime.
"Nuclear energy is emission free, and the new units are expected to provide reliable power throughout the region for the next 60-80 years. "
https://info.westinghousenuclear.com/news/westinghouse-celeb...
"...with a plant lifetime that can be readily extended to 80 years and beyond."
http://canes.mit.edu/overnight-capital-cost-next-ap1000