There's an option of overbuilding PV to fully accommodate cloudy days or the winter season (with extra PV going to waste on sunnier days). ...Versus making up the difference with stored energy.
You may be aware that large swaths of Finland, Norway and Sweden sees the sun set and not rise again for weeks or months. And during this period it’s also not uncommon to see -20 C or colder. Even worse/ the colder it is, it’s also typical with very little wind inside large stationary high pressures.
Then it’s months when the sun barely sets, as well as windy periods in spring and fall. Using wind/sun only requires storage not just between night and day but between seasons. Luckily in these particular regions there is plenty of hydro so aren’t reliant on wind and sun only to be 100% renewable.
Whether this is better than local storage is another question.
[1] https://research.tuni.fi/uploads/2019/05/0a103135-p086568.pn...
But chemical battery prices are also dropping as quickly.
At present the BESS (battery energy storage system) industry is (to a first approximation) a sideline for vehicle battery makers.
Relaxing engineering constraints imposed by vehicle use[1] means BESS prices can drop further. This is happening as the BESS industry splits off from vehicle batteries.
Add vehicle battery swapping like Ample's[2] to an urban BESS, you have two businesses in one, that can follow supply availability exactly.
1. Structural strength, vibration resistance, performance at extreme high and low temperatures, high mass energy density, high power/mass ratio, tolerance for overdischarge being the obvious constraints that can be relaxed a bit.
See this video which suggests somewhere in the region of 5x overbuild of renewables being the least cost option.
If that is the future -- I hope it is :D
Then there will be hours of the day where electricity is practically free, if not actually free. So any mechanism for time shifting energy consumption might generate a nice buck.
I guess the competition is PV or wind + long distance transmission loss.
A bit of Googling suggests that the combined energy per second that reaches Earth from all the visible stars other than the Sun is around 0.0000002% of the amount from the Sun.
I recall reading that the area needed with current solar panel technology to power the entire US would be 10000 square miles. With starlight being 0.0000002% of sunlight, that suggests we'd need 5 trillion square miles of panels to get the same amount of energy at night.
The surface area of the earth is a little under 200 million square miles.
That suggests that it isn't possible to overbuild enough to work off of starlight.
It would be even worse on cloudy nights. On cloudy days you still get a significant amount of sunlight coming through, because the Sun is giving us so much more than we need. Not so with stars.
It would be interesting to see exactly what kind of overbuild would be required to "fully accommodate" for the winter season with PV in Finland.