Phase change materials are a useful way of storing large quantities of energy.
As for warm air rising being used for energy, have you heard of the solar chimney technology that the Spanish invented? It has a giant greenhouse feeding hot air to a central chimney with turbines in it.
The problem is that the rock cools down and is slow to get hot again. This mean you don't generate much power for a big hole.
Geothermal energy is rather a small source compared to other long term sources, and is only really sensible to use in volcanic locations. In these you can pass water across really hot rock that is in contact with magma.
Yes! Just dig a mine shaft to the Earth's core.
I suggest you do a feasibility study, based on calculations using real world data and relevant physics.
It would help clarify whether your idea had any practical potential. If you are not concerned about that, I can see you would not need to.
I throw ideas out. Perhaps someone will find one of them useful. Perhaps not. It doesn't really bother me either way.
Sometimes they are best thrown out. 
I would point out that all obvious ideas have been considered by other people. You have to be rather creative and well-informed to think of a genuinely new technology these days.
Chinese researchers have achieved a huge leap in the performance of organic photovoltaics to over 17% efficiency. Last I heard, no-one had got much above 10%.
This type of photovoltaic can be printed on thin plastic or even fabric, so could have great potential for low cost mass production.
[Do I think I could invent a new PV technology? No.]
I like solar, but it has some serious limitations. Namely, operation without sunshine is nearly impossible.
This is why you need storage.
Solar energy can be used to produce hydrogen, or more convenient to handle, ammonia. These can be used to store energy for any time.
Some technologies such as solar chimneys incorporate short term storage (these can produce energy at night as well as during the day!)
There are places which have almost continuous sunshine (eg Sahara, Australia, parts of US, Brazil etc) so what is need is a way of distributing this resource around the planet. Electricity is after all not an energy source but a distribution system, So a kind of world wide grid is not so far fetched. It would far easier to build this technology and infrastructure than pouring billions into nuclear fusion (not fission) which always seems to be a pipe dream - or rather a tokamak dream. I'm quite happy to be proved wrong about about fusion but it always seems like an expensive way to go.
The sun generates enough energy on the surface of the earth in 90 minutes to meet global demand for an entire year.
s23bog, you are imaginative, but you seem to view wild imaginings as fact. How about checking them, if you have any wish to avoid being wrong? It sounds like great science fiction with a 19th century flavour, though!
The reason this is not at all easy is you need a gravitational gradient. On Earth, you can harness gravity through tidal power. In this case, the gradient acts on the entire oceans across the diameter of the Earth, which means it adds up to a useful amount of power. The total amount of power dissipated by tides in the oceans has been calculated: it is about 3 or 4 Terawatts across the entire Earth (much smaller than total human energy consumption).
By contrast, 174,000 Terawatts of solar energy falls on the Earth's atmosphere, most of it reaching the surface.
It is not (just) how many electrons pass through a socket that determines how much power you use. For example, if you attached an ideal capacitor to your electricity supply, electrons would flow in and out of it without consuming any power. By contrast, if you attached a resistor (say a one bar electric fire), the flow of electrons in and out of it would cost you money. What matters is that you extract energy.
Not all opinions are equal.
Well, the difference between attaching a capacitor (an ideal one) or a resistor to your power supply is that the power station does not have to use any energy to power the capacity, because it does not dissipate any energy.
Okay ... what are all the known wavelengths of light, sound, or other that are emitted by the sun? Which wavelengths are most useful for being converted to power?
Solar radiation peaks in the middle of the visible region. Most of the energy is in this region. The blue and green light is most useful for PV, the whole visible spectrum plus infrared is useful for solar concentrators. You may have missed my point that the total solar resource is about 50,000 as big as the total tidal resource.
For mining operations, there is going to come a point where utilizing the energy from the earth's core becomes a necessity. For a place like Hawaii, there has to be a way to produce usable energy from lava floes, or otherwise active volcanoes.
Again, renewable fuels (not to mention electricity sent down cables) are fine.
To utilise geothermal energy requires not just heat, but a reservoir of cold temperature and a way to transfer heat between them (usually liquid).