AI for real-time fusion plasma behavior prediction and manipulation

The amount of hydrogen contained in a medium-sized planet like Earth is extremely small in comparison with the amount of hydrogen contained in a star.

The amount of energy that can be produced by fusion per deuteron is smaller than the amount of energy that is produced in stars per proton.

With all these factors multiplied, the amount of energy that could be obtained from all the deuterium contained in Earth is many orders of magnitude smaller than the energy produced by the Sun or by any other star.

Moreover, the energy obtained from fusion could never exceed a very small fraction of the energy received by Earth from the Sun as light, otherwise it would lead to a catastrophic warming of the Earth.

Nuclear fusion reactors are not really useful for solving Earth's energy problems. They could have a crucial importance only for the exploration of the Solar System and for providing energy for human bases established on Moon, Mars or other outer planets.

For Earth the only problems worth solving are how to make better batteries, including very large capacity stationary batteries, how to make other large capacity energy storage devices, e.g. thermal devices, and how to improve the energy efficiency of the methods used to synthesize hydrocarbons from carbon dioxide and water.

Making hydrocarbons at large scale from carbon dioxide would be the best way to sequester carbon dioxide, offering the choice between just storing the carbon in safe products (paraffin like) and using a part of the synthesized hydrocarbons for generating energy in a carbon-neutral way.

On earth, there is an estimated 4.85×10e13 tonnes of deuterium; the energy density is 3.4x10e14 J/kg, giving a total yield of 1.649e31 joules. If you deleted the sun, this would be sufficient to maintain the current temperature of the Earth for ~9.5 million years: https://www.wolframalpha.com/input?i=%281.649×10%5E31+joules...

At "merely" the level of current human power consumption, this will last about 43 times longer than C3-photosynthesis, about 26 times longer than the oceans, about 5 times longer than before Andromeda merges with the Milky Way, and 6-3 times longer than when the Earth is currently expected to be absorbed into the outer envelope of the sun as it enters the Red Giant phase: https://www.wolframalpha.com/input?i=%281.649×10%5E31+joules...

Even if the sources I read giving those estimates are off by a factor of 10, deuterium alone, from earth alone, used as a total replacement for the sun, would still last longer than our species is likely to last before even natural evolution would have us speciate.

In the hypothetical future where we had a useful fusion reactor, the gas giants become harvestable, so the fact they're not on earth is unimportant. Likewise, on this timescale, every star in the nearest several galaxies — indeed, even absent novel technology and "merely"(!) massively scaling up what we've already invented, we already 'know'* how to get to places so far away that cosmic expansion is what would prevent a return trip.

As I said, it's technically correct that it is a finite resource. All I'm saying is that this is not a useful point on the scale at which we operate.

I expect it will be a useful point when we're star-lifting, but not now.

> Nuclear fusion reactors are not really useful for solving Earth's energy problems. They could have a crucial importance only for the exploration of the Solar System and for providing energy for human bases established on Moon, Mars or other outer planets.

I agree, however I also hope nobody makes a convenient cheap fusion reactor due to the proliferation impact of an affordable switchable source of neutron radiation.

> For Earth the only problems worth solving are how to make better batteries, including very large capacity stationary batteries, how to make other large capacity energy storage devices, e.g. thermal devices, and how to improve the energy efficiency of the methods used to synthesize hydrocarbons from carbon dioxide and water.

FWIW, I think that — if only we could cooperate better — a global power grid would be both cheaper and better than stationary batteries. Even just made from aluminium, never mind superconductors (and yes, I've done the maths). But we'd still need mobile batteries for transport, so that's fine.

The cheap abundance of PV power even today means I don't think we need to care much about making hydrogen electrolysis more joule-efficient.

> Making hydrocarbons at large scale from carbon dioxide would be the best way to sequester carbon dioxide, offering the choice between just storing the carbon in safe products (paraffin like) and using a part of the synthesized hydrocarbons for generating energy in a carbon-neutral way.

I suspect that carbon sequestration is unlikely to be a great win: there's a very narrow window close to zero loss/profit where on the loss side it's still cheap enough that people do it because it's a vote winner and on the profit side where it's not so profitable that people break photosynthesis a few hundred million years before natural processes do it.

* in the sense that Jules Verne "knew" how to get to the moon: the maths wasn't wrong, but the engineering was only good enough for a story

You can't pick it up with a forklift to put it inside the sealed container. That would make the forklift (and its operator) radioactive. You can only use a forklift after it's already within the sealed container. See for instance this real-life video (shot on a nuclear power station in my country), which shows used nuclear fuel rods being put inside a sealed container for long-term storage: https://www.youtube.com/watch?v=7X5K46ALdD0

Fortran compilers had more than 40 years to become pretty good at generating efficient code; they can make assumptions that are not possible in C (for example, no aliasing) to do so. Besides, most compilers already can do vectorization and autoparallelisation with multithreading, coarrays, and/or openMP, which can be offloaded to a GPU.