4 ms·
I agree that there is a need for reliably energy producers in standby to satisfy energy demand at every point in time. Using those for the base-load, as you su
by static_noise 12y ago
I agree that there is a need for reliably energy producers in standby to satisfy energy demand at every point in time.
Using those for the base-load, as you suggest, however does not make sense. The base load has to be supplied by renewables and nuclear energy. On top of that a quick-to-regulate energy source is needed. Nuclear only really makes sense for the base load as it best runs full power all day. About 50% Nuclear plus coal plus a bit gas is a good idea for a stable grid but not so good CO2-wise. Mixing a high percentage of nuclear with wind and solar does not make too much sense. Then you either need to throttle nuclear power or add another source many times. That leaves coal, oil, gas or hydropower for the grid security. The cleanest here is hydropower where additional turbines are installed but only activated when the demand is high. Using excess energy to pump water back up in the storage lake is also a possibility but more expensive than using what's already there.
An alternative to alleviate the strain on the grid is to regulate consumption. This is best done by having some large consumers getting cheap electricity but having to shut off when the grid demands it. Another idea is to use price regulation and the "smart grid". E.g. electric cars shifting their overnight charging time to adjust the demand with the supply. We'll have to see how far the "smart grid" takes us. When a large percentage of cars are electric cars with batteries it might make sense.