I have not followed all EPR nuclear reactor project, but it seems some project have a hard time to deliver. So is there somewhere an example of catastrophic failure ? or on the contrary, wind/solar project tend to be completed on time, with the expected cost for the provider ? Naively I would think that latter kind of project are much lower risks.
A 500 megawatt solar project is, approximately, 100 5-megawatt solar projects tiled together and sharing a transmission line. There are large parts of the project that can be built in parallel without scheduling dependencies. There are multiple vendors available for each equipment component (racking, tracking, inverters, modules) and there's a lot of mix-and-match flexibility. You don't have to buy Panasonic inverters and monitoring software just because you went with Panasonic modules. Large projects usually start generating power before their official commission date, because you can energize completed subsections as soon as they're connected to transmission.
Nuclear projects are pretty much the complete opposite. There are tightly coupled dependencies. A delay in one component (like a coolant pump) means that the entire project timeline slips. You can't play mix-and-match with reactor components -- no Westinghouse core plus Mitsubishi containment system. Unlike a large solar farm, a large reactor can't start generating partial power when the project is only 50% (or even 85%) complete, so financing weighs more heavily on the project economics.
Reactors that are up and running are great, and I deplore their premature retirement. But building new reactors is a slow process where the payoff comes only at the end.
Nuclear (from what I understand) usually has both far higher bid/pitched costs, plus an agreement to pay those over a longer period. Plus they usually come with finance or arranged finance.
The risk is in the building, but once built they are (in theory) a cash cow. But govt seem to find all sorts of ways to subsidise them - offering the clean up etc. P
