TLDR connecting far-flung regions into one huge grid can mitigate the intermittency of renewables. High-voltage direct current is a good way to do this.
Is this really more efficient than, say, building a big plant that does electrolysers or a bunch of giant flywheels to store excess energy? Or simply optimizing time-of-use to correspond when productivity is highest?
I haven't seen a comparison, but wikipedia claims HVDC loses only 3% per 1000 km.
https://en.wikipedia.org/wiki/Grid_energy_storage#Economics
The Atlantic Ocean is like 3–5000 kilometers wide. Over 5000 km you'd lose around 14%.
I mean, that's great. But what are the capital costs relative to a local power repository?
The 1,208 km euro-asia subsea power cable will carry 2000 MW of power and cost $745,000 per km, $900 million total
For comparison, this 2013 80% efficient scalable flywheel design costs around $1333 per kW. Handling 2000 MW would cost $1333 x 1000 x 2000 = $2.7 billion
The costs are comparable, but the 1,208 km euro-asia HVDC cable wins unless I fucked up my math or misunderstood how flywheel cost works
The length where flywheels overtake cables is $2.7 billion / $745,000 per km ≈ 3600 km, so flywheels might beat a trans-Atlantic cable, depending on the route
electrolyzers consume about 44kwh per kg H2 generated, and solar ain't efficient either:200w capacity per m²..