While I don't think this is entirely an answer I do agree with the general idea--utility infrastructure costs are based on peak load, not on average load. Solar cuts fuel use, it doesn't cut infrastructure.

I would be more inclined towards a system that is weighted towards when their system is under high load. You draw a lot of power when their load is low, it's not going to matter. You draw a lot of power when their load is high, it matters.

Many customers need costs to be predictable and bounded. So a cost structure that depends on external factors is very rough on those at the margins. You can encourage many people to move to rate plans that depend on external factors, but not all.

And I'm not sure how your plan addresses the problem. Suppose I install enough panels and batteries for 48h of power. I only draw from the grid when there are 3 cloudy days in a row. If I pay by kWh, I don't really care how much it costs per kWh since I use so few of them. Yet I cost the grid operator just about as much as a normal customer since they need to maintain the capacity for that 3rd cloudy day.