> This was done on purpose. There is nothing inherent in the technology that makes it expensive. "Generate heat to generate steam to generate electricity" is fundamentally the same for nuclear and coal or natural gas, so the only explanation for the former having a significantly higher cost is regulatory, which was intentional and could be different given a different intent.
Nuclear is vastly more complicated and the ultimate failure condition that everyone worries about can cost hundreds of billions to clean up.
That results in a lot of engineers and administrators sitting around tables and dreaming up possible ways things could fail (particularly with humans in the loop) and dreaming up ways to mitigate them. The designs get more complicated and expensive over time because nobody wants to say 'no' to mitigating those risks.
The engineering of the plant also needs to be built to exacting standards and none of the pipes or concrete are what you find slapped on buildings in your local neighborhood.
Contrast that with a solar or wind plant or even a coil or nat gas plant and nobody cares that much about the engineering. If something breaks you just shut it down and fix it and maybe retrofit the design (bit more care needs to be taken in the case of a coal or nat gas plant since you don't want it blowing up but nothing out of the ordinary for any refinery or other industrial plant). With a nuclear plant you can't just pull the plug and shut it off to fix it without creating a 40-year long cleanup disaster.
You're explaining why they cost more to design, not why they cost more to manufacture. The kind of redundancy you need for this doesn't require an excessive amount of raw materials. The cost of a diesel generator that can run coolant pumps, or five of them, isn't a meaningful percentage of the construction cost. A large proportion of the plant isn't any more safety critical than it is in any other kind of turbine-based generating plant, because it's doing the same thing in the same way.
It's also working against one of the most effective design criteria for ensuring safety: Simplicity. Things with fewer moving parts fail less. But they also cost less. It shouldn't cost more. Something's being done wrong.
And if your argument is that nuclear plants are required to have higher safety standards than anything else does even for doing the exact same thing with the exact same risk as it is in some other kind of facility, yes, that's the point -- that's why it's a regulatory problem. Uneven application of standards. Either one of them is too loose or one of them is too tight, but they shouldn't be different.
They're different because one of them has a risk of a meltdown with hundreds of billions of dollars of damage. That's the asymmetry and why a nuclear plant costs more than a refinery.
Nuclear is vastly more complicated and the ultimate failure condition that everyone worries about can cost hundreds of billions to clean up.
That results in a lot of engineers and administrators sitting around tables and dreaming up possible ways things could fail (particularly with humans in the loop) and dreaming up ways to mitigate them. The designs get more complicated and expensive over time because nobody wants to say 'no' to mitigating those risks.
The engineering of the plant also needs to be built to exacting standards and none of the pipes or concrete are what you find slapped on buildings in your local neighborhood.
Contrast that with a solar or wind plant or even a coil or nat gas plant and nobody cares that much about the engineering. If something breaks you just shut it down and fix it and maybe retrofit the design (bit more care needs to be taken in the case of a coal or nat gas plant since you don't want it blowing up but nothing out of the ordinary for any refinery or other industrial plant). With a nuclear plant you can't just pull the plug and shut it off to fix it without creating a 40-year long cleanup disaster.