In this month's Wired
front page story about electric cars, Roth brings up one of the problems with wind and solar power:
DONG [Denmark's largest utility] makes a higher portion of energy from wind—18 percent—than any other power company in the world. Danish politicians want to see that figure doubled, which is good and green but completely impractical: Some days the wind blows, and some days it doesn't. Banking wind energy is expensive and inefficient—DONG would have to buy fields of batteries.
Turns out there is a neat solution to this problem: use hydroelectric plants' reservoirs to accumulate potential energy of water. It is possible to control the amount of water that goes to turbines and whether the level in reservoirs rises or falls. By tuning hydroelectric energy production in response to supply and demand of the grid, you can "smooth out" short- and medium-term fluctuations with near-perfect efficiency.
They probably don't have many hydroelectric plants in Denmark, but Soviet Union has built a bunch of them. In 1991, I wrote Fortran code to simulate USSR's power grid and optimize the use of power plants and transmission lines. My grandfather, Anatoly Zeyliger, got me the gig: he worked on the energy infrastructure. This might be one of the domains where free market principles should be applied with caution, and central planning is not such a bad idea: global optimizations lead to huge savings. My grandfather declared California's deregulation plan "a bad mistake" long before Enron's traders exploited loopholes to rip off taxpayers.
Here's a physics problem that bugged me for the last twenty years. Two people have to carry a large object up a narrow staircase. (Twenty years ago, that was a tub full of firewood for the furnace on the second floor of our dacha. Today, it's a big screen TV for my grandma.) Whose job is more difficult: the person who goes first and holds the top end of the object, or the second person who carries the bottom end?
