[somenameforme]

I see a couple of issues here. The first is that creating, maintaining, and operating a nuclear facility on the Moon would almost certainly be far more challenging than "just" maintaining a couple of weeks of power storage. But that kind of implicitly leads into the other issue. That is that you only really need the scale of power that nuclear can offer once you've already substantially industrialized the Moon. For some simple habs and research areas, even something as small/simple as a radioisotype generator [1] would be more than fine.

Beyond all this, I meant novel when I said novel. The regular extremes of heat and cold offer all sorts of interesting ideas. You've got room for predictable and endless convection on basically an arbitrarily large scale there. There is certainly going to be some clever way to exploit this in a novel fashion.

[1] - https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_ge...