We need grid-scale energy storage.
Right now we have about 30 GW worth of it on a grid with 1200 GW of generative capacity.
Until that changes, every solar and wind installation is going to come paired with a gas peaker plant.
The future isn't solar/wind + batteries, at least not much beyond the point where the overbuild requirements are such that one is building more and more generation and storage to compensate for smaller and smaller windows of time. Which for most places is probably around the point where the total solar and wind capacity is roughly equal to the peak production average times a small multiple less than 2.
I always point at texas, and which is again in the news, because they are discovering that solar quits about two hours before the peak demand falloff, and they are literally paying people not to consume power. And this is only really a problem because no one wants to invest in NG plants that sit idle most of the time, and the wind generation has dropped down to 20% or so of its nameplate installed capacity over the past couple weeks. Leaving literally nothing to make up the shortfall. So there is going to be a huge rush to install enough batteries to make up that 2 hour window for 1/2 the year, then it will stop because its not economical to have them providing power outside of that window.
Texas doing its own thing is, sadly, evidence against my personal favourite solution: a global power grid would, from a pure engineering perspective, be totally viable, and would mean you don't need any storage for stationary use — sun always shines somewhere on the planet, an a square metre cross section of aluminium is close to just one Ohm per 40,000 km, and that's only a few year's global aluminium for something that would easily last a century.
Unfortunately, Texas isn't even willing to join up with New Mexico, so this is unlikely to come to pass.
The wildly misleading number is about what China spends on coal in a couple of years: a few hundred billion.
Also, at 40,000 km you don't need two — that's all the way around the planet and back to the start.
Heat loss obviously depends on current and voltage choices; reasonable numbers say this still works out fine, given how much cheaper PV is compared to both storage and other production.
However, despite all that: if we can't even take baby steps like convincing Texas to play nice with their neighbouring states, something like this is only plausible on planets with a single world government, not Earth.
If they run just a couple of days a year, they could be fueled with green hydrogen with little effect on the overall cost of the system. Fossil fuels are not needed here.
> Until that changes, every solar and wind installation is going to come paired with a gas peaker plant.
Sure; and as the necessary gas plants are mostly already all there, we can start by doing nothing, before gradually just switching them off as we build out the storage.
(The storage doesn't have to be batteries: gravity storage is cheap; disappointingly, hydrogen electrolysis doesn't exist in meaningful quantities yet, but it is no more mysterious magic than batteries; and there are more besides).
Only way to be sure we don't get them is to invent something better.
I do like the aesthetics of hydroelectric dams: they also need concrete, though proportionally less.
I think some people are working on eco-friendly cements that would make for eco-friendly concrete; we certainly need that too, regardless of how we solve power production.
We need some grid-scale energy storage. But having distributed storage is a welcome element as well, it makes the grid far more stable and reduces the impact of peak production times as well as the day/night cycle.
Anyway, on a planetary scale energy is more of an engineering challenge than a serious problem, the issues all stem from the hundreds of local fiefdoms that all vie for control of a slice of the pie.