Comment on AI experts return from China stunned: The U.S. grid is so weak, the race may already be over

• p03locke@lemmy.dbzer0.com ⁨19⁩ ⁨hours⁩ ago

  No one ever warned us that our energy needs have been climbing steadily for decades, and we need to stop being afraid of nuclear power? Really?

  What rock have you been hide in?

  source

  • Rivalarrival@lemmy.today ⁨16⁩ ⁨hours⁩ ago

    Nuclear and solar have competing problems. Nuclear is a baseload generator. It can’t ramp up or down fast enough to meet the daily demand curve; it needs a steady, stable load, 24/7. The steadier and stabler the load, the better. If the load drops off overnight, nuclear has to dial back its continuous output to match that trough. And again: It can’t ramp up and down fast enough to match demand, so it just has to stay at the lower “trough” level, with the remainder made up by various types of “peaker” plants.

    To make nuclear as efficient at possible, we need to drive consumption to that trough. We have to increase overnight demand as high as possible, to minimize our reliance on inefficient peaker plants.

    Now, look at solar. Solar stops generating overnight. Solar can’t possibly meet overnight demand without storage, and grid-scale storage solutions are fundamentally limited. To make solar as effective and efficient as possible, we have to move as much demand to daylight hours as possible, where it can be met directly by solar generation, without storage.

    The two technologies require opposing demand incentives. Making one more efficient necessarily makes the other less. Whichever choice we make here, the other one is relegated to a limited, auxiliary role in generation, and can never reach its full potential.

    source

    • WoodScientist@lemmy.world ⁨3⁩ ⁨hours⁩ ago

      The solution is to just spam solar panels. Solar power is getting so comically cheap, that this is the solution we’ll likely use. Have enough storage for overnight. Then spam so many solar panels that your grid can meet demand even on a cloudy day in winter. You have enough to meet demand at the lowest productivity point in the year. Then the rest of the year we have cheap hyper-abundant power.

      source

      • Rivalarrival@lemmy.today ⁨2⁩ ⁨hours⁩ ago

        Exactly. That is exactly what we need to do.

        Then the rest of the year we have cheap hyper-abundant power.

        Ideally, yes. But, what is actually happening is that near the summer solstice, generation rates aren’t “cheap”. They are negative. We are putting so much power on the grid that generation companies are paying for people to take it off during ideal generation conditions.

        That is a big fucking problem. Negative rates mean we stop “spamming” solar panels long before we have enough to meet winter demand.

        The solution to that problem is 3-season industries. Major industrial consumers that only operate from spring through autumn, soaking up the excess power, then going offline, shedding their excessive load for the winter.

        source
    • Diurnambule@jlai.lu ⁨3⁩ ⁨hours⁩ ago

      Batteries, mass of batteries.

      source

      • Rivalarrival@lemmy.today ⁨3⁩ ⁨hours⁩ ago

        Not feasible.

        It’s barely feasible to use pumped storage for solar to match the daily demand curve in some small areas. Grid scale storage cannot be feasibly scaled to serve our current overnight power needs. But the daily demand curve isn’t the problem.

        The real problem is the seasonal variation.

        For solar to be effective, it needs to be able to meet our winter demand with our winter sunlight. 9 hours of low-angle sunlight under largely overcast conditions. That means we need a lot of solar panels, to get sufficient power from these suboptimal conditions.

        Now, take that same number of solar panels, and give them the 15 hours of high-angle sunlight under largely clear skies that we have during the summer. When we do this, we have so much power pushed on to the grid that the price of electricity actually goes negative. They literally have to pay people to take it.

        There isn’t enough lithium in the world to make the batteries we would need to balance seasonal variation. There isn’t enough land on the planet to support pumped storage facilities that could balance seasonal variation.

        We need demand shaping to make solar feasible as our primary energy source, which means driving our heaviest loads to daytime, away from the dark. (This is the exact opposite of what we need to do for nuclear, coal, and other baseload generation.)

        We also need 3-season industries that can soak up excess production in spring, summer, and autumn, while going offline and shedding their loads during winter.

        source