OK, Thanks. I read the first one. It seems to say what you said: storage cost of < 1cents/kWh. So I am having a hard time reconciling the NREL battery storage cost with this 2018 article. Hopefully the author didn't make any mistake in coming up with that 1c/kWh.
There is a link for 'quietly reported', but that seems like author's document and costs are in different units (MWh vs. kW-mo).
The total project cost doesn't explicitly say the battery capacity in the project, so 1c/kWh is still little doubtful, imo.
How much was the big battery in Australia? I think it was $100M+ for some 129 MWh, which is much much higher than 1c/kWh storage with charge&discharge.
Here is another report from EIA published Feb 2019. This shows a somewhat different picture of cost of new power plants.
Just solar is almost same as new CC plants. I don't fully understand the various tables here. Will try some other time to understand what exactly these LCOE and LACE mean.
https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf
The 1 cent is over the life of the battery system and it's not built 1 kWh battery to 1 kWh solar, just a fraction. CAISO has nice charts that can tell you how much the system would get used over the year, by month. Scroll down to battery and you will see how the battery is used day by day. Battery is currently deployed to act as a grid balancer, more like a peaker plant except it's way better than a peaker plant in meeting instantaneous demand, thus no need to overbuild and overproduce.
California ISO - Supply
Based on Time of use rates, if I were to buy a Powerwall, I would only expect to use it 4 mo out of the year and only during the peak rate period. So, my Powerwall should last a very long time, as in several decades.