3 ms·
Capacity over build is already factored into LCOE for renewables, right there in the definition. Transmission lines are not but comparing how many HVDC lines y
by Gwypaas 4y ago
Capacity over build is already factored into LCOE for renewables, right there in the definition.
Transmission lines are not but comparing how many HVDC lines you can get for the difference between renewables and a single nuclear reactor is laughable. This is not even mentioning the cost that has gone into building the current grid in a hub and spoke model around big centralized producers.
Not that we need to throw it away, but it is worth mentioning when talking nuclear. Especially new builds.
Power 2 X makes storage solved at costs lower or in line to nuclear. Not that anyone would accept those costs.
The real interesting research which is currently happening is around grid forming inverts. Being able create a country wide grid without any synchronous producers or coupling.
- acidburnNSA 4y ago> Capacity over build is already factored into LCOE for renewables, right there in the definition. The way people argue about it, no it absolutely not. CAISO generates about 25,000 MWd in a day. If you just use LCOE you would say, ok let's take the LCOE in $/MWh, convert to $/MWd and multiply them for each energy source. Wow look how cheap 100% wind and solar are! But oops they totally missed the fact that oops in Winter they have to build at least 4x capacity to fill up the shockingly large amount of chemical batteries (hundreds of skyscrapers full) and wires during the windy/sunny periods, and then rebuild all of this stuff every 25 years. Now it looks very expensive and crazy. The 92% capacity factor nuclear plant running for 80 years is looking good. Wind and solar alone are the cheapest forms of energy as tents are the cheapest form of housing. Many studies suggest that a good mixture of hydro, nuclear, wind, and solar are the ideal. Let's not rule out nuclear due to current market quirks.