Slashdot Mirror
Batteries To Store Wind Energy
Roland Piquepaille writes "Scientific American reports that Xcel Energy, a Minneapolis-based utility company, has started to test a new technology to store wind energy in batteries. The company is currently trying it in a 1,100 megawatt facility of wind turbines in Southern Minnesota. The company started this effort because 'the wind doesn't always blow and, even worse, it often blows strongest when people aren't using much electricity, like late at night.' It has received a $1 million grant from Minnesota's Renewable Development Fund and the energy plant should be operational (PDF) in the first quarter of 2009. If this project is successful, the utility expects to deploy many more energy plants before 2020 to avoid more polluting energy sources."
They do that. It's called Pumped-storage hydroelectricity.
Assuming concrete is reasonably environmentally friendly this would be a pretty clean solution.
Concrete has a massive carbon footprint. The calcination of lime releases a lot of CO2, on top of the fossil fuels used in manufacture and transport.
Oklahoma already has two hydo plants and one pumped storage plant. You don't need huge elevation changes, a few hundred feet will do.
Chernobyl 'not a wildlife haven' - BBC News
3. Given the high velocities - what will happen when they fly apart? Also, the gyroscopic effects they generate while spinning.
For a stationary plant, have it spin horizontally, and build it underground.
If it does suffer a catastrophic failure, loss of life and damage to surrounding infrastructure should be minimal
---
"I can't complain, but sometimes still do..." Joe Walsh
I did some calculations (yay!), and came up with the following: Raising the Empire State Building (365,000 tons of material) to the height of one meter would store a little shy of a megawatt hour of energy.
I'm imagining this weird future city where the buildings slowly rose and fell as energy was stored and withdrawn. It's a cool thought, but it seems that the engineering difficulties would be considerable, and the payoff not so much.
A system where water was stored at the top of a skyscraper might be more feasible (putting the weight a hundred times higher means you only need 1% of the material. You might be able to do something with water, or a block on a chain. But the storage payoff seems relatively small.
It might make more sense to deal with material that's already being lifted up and dropped down. Like integrating some sort of storage and release system for the water already being pumped to the top of skyscrapers. Given separate reservoirs for potable water and sewage, and some leeway about when to pump water in and release waste out, something might be arranged.
The calculation: 365000 tons * 907 kg/ton * 10 joules/kg * 1kWH / 3,600,000 joules. The 365000 tons figure is from this kid's site.
You want the truthiness? You can't handle the truthiness!
The weight of a flywheel is far less important than the weight. Energy of motion is a product of mass times velocity squared, so doubling the rate of rotation quadruples the amount of energy stored. The ideal for many applications is to find a material that is very light, but won't fly apart at high speeds. I remember reading about somebody trying something with carbon nanofibers, but that was a long while back.
The weight doesn't matter much for energy seepage either. A good flywheel will be suspended by magnets, so regardless of the weight, the friction due to weight is effectively zero. There is still air friction and electrical losses to deal with.
Getting energy into them isn't a huge obstacle either. I've scoured the web, and it looks like the people actually selling flywheels-as-UPS solutions are claiming 90% efficiency.
Something may be missing in my understanding here. The article claims that the battery backup for the wind farm costs about three million dollars per MWH, whereas the flywheel backup system I'm looking at right now claims that it can give you about 200kWH capacity for about $50,000. That's $250k per MWH installed, and very low maintenance costs.
Their claims could be overblown, or I could have my math wrong, or there is something I'm missing that makes the flywheels unsuitable for this application. There might also be a huge economic opportunity, but somehow I doubt it.
You want the truthiness? You can't handle the truthiness!
Flywheels are attractive for short-term peak power delivery. They have low failure rates and easy fault detection (if the wheel is intact and spinning at the required speed, you know how much energy is available).
For long term loads (hours) flywheels aren't competitive with lead-acid batteries, let alone more exotic types such as the NaS battery the article describes. For example, the Active Power CSDC-500 flywheel storage system supplies 50kW for 138 seconds = 1.92kWhr. The cabinet is 78" x 54" x 34" and it weighs just over 3 tons. Four long-term loads, a system with two 12V 100Ahr VRLA batteries would be 14" x 14" x 10" and weigh 140 lb.
A flywheel based system has nowhere near the energy density of a battery storage system. Peak power density is the flywheel's forte.
I am so sick of science writers who mess up the story because they don't understand the units of energy and power.
The article says the batteries store 7 megawatt hours. Fine.
Then it goes on to say "meaning the 20 batteries are capable of delivering roughly one megawatt of electricity almost instantaneously" WTF does that mean? Power, measured in megawatts is by definition an instantaneous unit. What's with "almost instantaneous". Also, the rate of discharge of a battery MW is unrelated to its storage capacity MWh, so the entire meaning of the sentence makes no sense.
Then the article says, "Over 100 megawatts of this technology [is] deployed throughout the world," Huh? Battery capacity is measured in megawatt-hours, not megawatts.
Then the article says, "costing roughly $3 million per megawatt" same thing. Battery cost must be proportional to megawatt-hours, not megawatts.
I suspect that their idea is to make a battery with 24 megawatt-hours of capacity able to deliver 1 megawatt of power uniformly for 24 hours, then say so.
Shame on Sciam writers and double shame on Sciam editors for not mastering such basic units in an article about energy.
How about a little economics. The article mentions two understandable numbers, an 11 MW wind plant, and 7 MWh of battery capacity. The combination of the two, allowing for wind variations during the day believably deliver 1 MW continuously to the grid. That's 24 MWh per day.
Now the batteries cost $3 million, and the wind generators cost $22 million. Total $25 million to deliver 1 MW of base load. That's $25 billion per GW.
The peak generating capacity of North America is about 750 GW. Let's say 250 GW when levelized to base load. Therefore, to supply 100% of that with wind and batteries would cost roughly $6.2 trillion dollars. Now Al Gore says, "No problem. We can do that in just 10 years." WTF is he thinking?
Even if we did spend $6.2 T, there will still be periods where not much wind blows for large regions for many weeks at a time. I live where it's cold, and I know that when it hits -30F, the wind is almost always still and the sky dark, and that it can stay like that for a couple of weeks. We therefore, need to double or triple the $6.2T plus more for transmission, to provide backup power sources, plus the means of delivering the energy over large distances.
Wind and solar are wonderful for up to 15-1-20% of the total grid generation and the cost of construction and operations dominate. More than that, and reliability and deliverability of the electric supply become dominant in the economic equation.
So if a reactor is built in the US to today's abbreviated containment standards, yes, that reactor would include a failure mode similar to what occurred at Chernobyl.
The failure mode at Chernobyl was very specific to the design of the RBMK-1000 and to it being near the end of the core life. The problems at Chernobyl were that it had a positive coolant void coefficient, the reactor was burning Plutonium (delayed neutron fraction of 0.2% versus 0.65% for 235U), the graphite moderator was not thermally coupled to the fuel or coolant, and last but not least, the scram rods increased reactivity at their initial portion of travel - the Chernobyl accident was triggered by an operator scram'ing the reactor.
American light water reactors were design explicitly to have a negative coolant void and temperature coefficient.
FWIW, I do have a degree in nuclear engineering.
A Shadeless room is a brighter room.
Put the flywheel in a permanently-sealed vacuum chamber.
no such thing. Not that it's a big deal. Commercial vac dewars and such have ports for pulling out excess atmosphere that seeps in anyway. you would use the same thing here. Wrap the outside of the chamber with LN2 pipes to cool the air inside so that is is less energetic and "falls" to the bottom of the chamber, then with the help of a turbo pump suck the chamber of all air present that is reasonable to get out.
-nB
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump