I know I'm not the only one here with an interest in this sort of subject - I'm more or less thinking aloud, and would be grateful to know what others think!
Scenario: we have 5.7kW solar PV on the roof. Despite the car being a PHEV, we still export 75% of the energy. We have an export meter so we get paid ~5p/kWh export; I pay ~15p/kWh to buy grid electricity. Across a whole year I export 4400kWh, whilst importing 3000kWh. I'm on the verge of switching the gas heating to a small ASHP and wet underfloor heating.
I'm considering buying a battery. This would be one of the newest types, using re-purposed EV batteries; I'm loking at an 8kWh model from Powervault. From an economic point of view it will pretty much break even over its lifetime. Assuming it runs smoothly for the predicted ten years, the saving from the difference in price for each kWh it saves me from importing will add up to its installation price, more or less.
What I'm mulling over is whether it's a good idea in terms of overall reduction of CO2. As it's made from recycled car batteries, I'm choosing to neglect the embedded CO2 in producing it - whilst this still won't be zero, it'll be small enough to ignore over a ten year period.
If I look just at my household CO2 it's a clear winner; it saves CO2-free energy from the PV and displaces grid electricity. Although I pay the premium for 100% green electricity, I know full well that reducing my demand reduces overall grid emissions, probably to the tune of the UK average of around 270g/kWh.
But: using my PV to charge my battery is simply depriving the grid of that contribution. What I am really achieving is to not supply the grid with energy when it's sunny, in return for not drawing from the grid when it's dark. I am still using the same amount of power (actually a little more, because the cycle efficiency of this battery is in the region of 90%, so a full charge-discharge actually uses about 10% more energy than otherwise).
What I can't answer, then, is whether this is worth it or not. If we had a magic 100% renewable grid, then clearly not. If we ever got to the point where the daytime demand was less than the renewable supply, and we had excess renewable energy to use, then clearly it would.
I fear the answer is probably that it's not a net saving of CO2 - all I am doing is removing power from a time of day when demand is high and the nastiest types of generation such as coal are more likely to be switched on, and saving demand in the night when it's likely that I'm just displacing CCGT. Or, at best, I'm just displacing daytime CCGT for nighttime CCGT, which doesn't really make any difference at all...