>And the optimization problem at this point in time is that -ideally- you want to have some amount of excess generation on your solar panels, then time shift that with batteries to a point in time when power is more expensive, and have some amount of excess capacity in batteries so you can also sell that power; versus the (amortized) cost of that extra capacity; and/or what will actually fit on your roof.
Yep, that's the ideal case - given you're in the region where you have reliable sunshine in the winter. It's not the case where I live, when we had total of 7 hours of sun in December 2023.
The idea would be to sell excess power in summer to offset the cost of buying power in winter.
For sure this won't work equally well everywhere, and I'm not sure if this would still work once practically everyone has solar; but right now it's plausible.
Okay, but the "excess power in summer" does not magically materialize in winter. The current systems where it works like that are accounting trick meant to solar power industry.
In reality, we have no realistic mechanism for long-term energy storage on a grid scale - we barely start to breach scale where storage can handle daily fluctuations - like in California. And it's not free - it's being costly affair.
You're looking at this from a very different angle than I am, I think. While I acknowledge that at the system level we still have a way to go, we're already at the point where individual homeowners can sell power in summer and use that money to offset a significant portion of their winter costs - if they play their cards right. This won't solve all our global or national energy challenges, but it can make financial sense at the individual household level.
Yep, that's the ideal case - given you're in the region where you have reliable sunshine in the winter. It's not the case where I live, when we had total of 7 hours of sun in December 2023.