3 ms·

> You can replace a significant proportion of the grid with intermittent generation without any trouble. Yes, in some locations. That’s a no brainer and where

by klabb3 3y ago

> You can replace a significant proportion of the grid with intermittent generation without any trouble. Yes, in some locations. That’s a no brainer and where it’s economical it’s already happening. But unpredictable supply particularly needs on-demand energy. Nuclear can’t respond quickly, so if you don’t have hydro you’re likely left with gas. > Use whatever you want to charge electric car batteries because they're batteries and most cars are parked >90% of the time. Agreed. We have some way to go with universal infrastructure support for that though. > You can can't shift 100% of the load based on time of day through pricing, but you can do a non-trivial amount. Possibly. The major sources today are heating, AC, fridges. Cooking will come up there if you remove gas. In either case, industrial and commercial are ~60% of use and how do we move that significantly out of peak hours?

adrianN 3y ago
For short term unpredictable demand it is likely that batteries will outcompete gas in the next couple of years.
AnthonyMouse 3y ago
> But unpredictable supply particularly needs on-demand energy. Nuclear can’t respond quickly, so if you don’t have hydro you’re likely left with gas. Nuclear is baseload, but baseload offsets intermittency. If you're using solar you've got no generation at night. Suppose you do all the load shifting to the day that you reasonably can and you still need 600 GWh of power for the 12 hours between sunset and sunrise. Well, add 40GW of nuclear and now you only need 120 GWh. You've cut back the amount of gas or batteries you need by quite a lot, but you still have some, which is enough to respond to short-term demand changes. And the nuclear plant generates during the day too, which isn't as valuable but is still not zero and contributes to paying its costs. > We have some way to go with universal infrastructure support for that though. At some point there was a fear that this was chicken and egg: No one buys electric cars because there are no chargers so no one builds chargers. There are enough chargers now that it's viable for a significant proportion of people to buy an electric car. At that point people will just build more chargers in proportion to how many cars need them and there isn't really anything to worry about. And the power can be generated from renewables without having to worry about intermittency because you're charging a battery which is typically stationary for most of the day. > The major sources today are heating, AC, fridges. Cooking will come up there if you remove gas. AC is fine, aligns well with solar. Heating is the hard one, but heating is approximately baseload. The best candidate for that seems to be nuclear -- which can also potentially provide cogeneration for buildings in reasonable distance from the plant. > In either case, industrial and commercial are ~60% of use and how do we move that significantly out of peak hours? Businesses disproportionately operate 9-5, which is daylight hours. Many industrial operations operate 24H because the cost of paying people a premium to work second and third shift is less than the cost of building three separate factories that only operate 9-5. Now they'll each have to decide if it's also worth paying the cost of energy storage to operate at night.

by klabb3 3y ago

adrianN 3y ago

For short term unpredictable demand it is likely that batteries will outcompete gas in the next couple of years.

AnthonyMouse 3y ago

> But unpredictable supply particularly needs on-demand energy. Nuclear can’t respond quickly, so if you don’t have hydro you’re likely left with gas. Nuclear is baseload, but baseload offsets intermittency. If you're using solar you've got no generation at night. Suppose you do all the load shifting to the day that you reasonably can and you still need 600 GWh of power for the 12 hours between sunset and sunrise. Well, add 40GW of nuclear and now you only need 120 GWh. You've cut back the amount of gas or batteries you need by quite a lot, but you still have some, which is enough to respond to short-term demand changes. And the nuclear plant generates during the day too, which isn't as valuable but is still not zero and contributes to paying its costs. > We have some way to go with universal infrastructure support for that though. At some point there was a fear that this was chicken and egg: No one buys electric cars because there are no chargers so no one builds chargers. There are enough chargers now that it's viable for a significant proportion of people to buy an electric car. At that point people will just build more chargers in proportion to how many cars need them and there isn't really anything to worry about. And the power can be generated from renewables without having to worry about intermittency because you're charging a battery which is typically stationary for most of the day. > The major sources today are heating, AC, fridges. Cooking will come up there if you remove gas. AC is fine, aligns well with solar. Heating is the hard one, but heating is approximately baseload. The best candidate for that seems to be nuclear -- which can also potentially provide cogeneration for buildings in reasonable distance from the plant. > In either case, industrial and commercial are ~60% of use and how do we move that significantly out of peak hours? Businesses disproportionately operate 9-5, which is daylight hours. Many industrial operations operate 24H because the cost of paying people a premium to work second and third shift is less than the cost of building three separate factories that only operate 9-5. Now they'll each have to decide if it's also worth paying the cost of energy storage to operate at night.