3 ms·
I think the minimum remains at 0 - the no-wind today condition.
by koheripbal 4y ago
I think the minimum remains at 0 - the no-wind today condition.
- im3w1l 4y agoThis made me try to imagine solutions to the no-wind condition, whether there is any control strategy that can work for it. And I laughed at the idea of one wind mill blowing at another to get a net gain of energy. But then I thought, is that really such an absurdity..? What if.. what if you can expend energy to "suck in" a nearby storm for instance? I don't know how viable it would be, but at least it doesn't seem to obviously break physical laws.
- hinkley 4y agoStored hydro is basically this. I’d like to see stored hydro that was a little more environmentally friendly. Perhaps a series of floodgates where one area is mostly dry and another mostly wet, rather than everything being intermittently wet all year.
- sgc 4y agoI get it. You want to store energy in man-made tornadoes, then discharge them slowly when wind calms down. So the optimal layout is circular. By adjusting the turbine, they can either feed into or pull out of the cyclone. Genius!
- HPsquared 4y agoI think it's more like cloud seeding, trying to redirect a weather event (which is happening anyway) to a more fortuitous location.
- im3w1l 4y agoWhile an interesting thought, it's not at all what I had in mind. I was imagine stealing the wind from the neighboring town. Let me draw a figure wind | x x | | x x v wind \ x-> x-> \ x x
- hinkley 4y agoNo wind all day is quite a bit different from spotty wind. Any day with a little wind slows your withdrawal rate. And while you might not be able to decommission a peaker plant, one of the ways power gets around emissions limits is that the pollution is annualized. You can reduce the emissions of a plant by half of you can turn it off part of the year, so where possible they spin up the better plants first. Anything that keeps their next worse plant offline more helps the rest of us.