In winter time, the sun is low in the sky and not up for that long. There's no way it would be providing energy 80% of the day. More like 25%. And with the sun low in the sky, and the road isn't tilted to face the sun, it's hard to see how it would generate much.

First, the 80% number is a number I pulled out of a hat. Second, why does it have to be constrained to that day? Why not (effectively) bank it in the summer and use it in the winter?

There's the sticky widget. Power storage is a huge problem at this point that we don't have a really good solution to.

Well, we do. It's just that the methods involve a lot of loss.

Plants store solar energy as hydrocarbons. Further reformed into coal, oil, and gas, that's what we use today.

It's possible to synthesize hydrocarbons from seawater at a net efficiency likely around 50%. Hydrogen electrolysis is the primary energy cost at 60% energy delivered on input.

Run that from solar cells at 20% efficiency and you've got a net delivery of 10% of input solar as stored hydrocarbons. Mind that you're only going to extract about 30-40% of that as useful work or energy (electric generation is about 38% efficient in fuel conversion, cars around 10-20%, IC engines generally about 30%).

But the neat thing about hydrocarbons is that once you've got them, they can store for a long time. Hell, we've got proofs-of-concept showing 300-500 million year viability.

http://redd.it/22k71x

There are other options as well. Some are far more responsive (flywheel kinetic storage, pumped hydro), some far more efficient in energy input/output), others have varying degrees of suitability (CAES, thermal energy storage, liquid metal/molten salt batteries). I'm actually starting to see this as a potentially solved problem.