From the American Institute of Physics
Using fluctuating wind power
Incorporating wind power into existing power grids is challenging because fluctuating wind speed and direction means turbines generate power inconsistently. Coupled with customers’ varying power demand, many wind-farm managers end up wasting power-generation capacity and limiting the service life of turbines through active control – including fully stopping turbines – in order to avoid any possible damage to the power grid from spikes in supply. In a paper published in the American Institute of Physics’ Journal of Renewable and Sustainable Energy, researchers propose a new strategy to optimize power-generation efficiency and so better control wind farms.
The new strategy is based on continuous predictions of how fluctuating winds affect each turbine’s maximum generation capacity. It also incorporates factors missing in other wind-farm control strategies, including differing power generation between turbines, actual fluctuations in power generation capacity, errors in prediction, communication disruptions preventing active control, and even turbines without the capacity for continuous active control. To demonstrate the feasibility of the new strategy, the researchers compared their predictions to raw data from a single wind turbine. The team then further refined their calculations and simulated a control operation with data from a wind farm of 33 turbines.
The results suggest that wind-farm managers can improve their power-generation efficiency with the new strategy. However, the researchers caution that before implementing the strategy, each wind-farm manager should adjust the underlying parameters – such as how often to adjust each turbine’s speed – based on local conditions.
Article: “An Active Power Control Strategy for Wind Farm Based on Predictions of Wind Turbine’s Maximum Generation Capacity,” is published in the Journal of Renewable and Sustainable Energy.
Link: http://jrse.aip.org/resource/1/jrsebh/v5/i1/p013121_s1
Authors: Dewei Liu (1), Jianbo Guo (1), Yuehui Huang (1), Weisheng Wang (1)
(1) China Electric Power Research Institute
Related articles
- Rethinking wind power – Harvard study shows it to be overestimated (wattsupwiththat.com)
- Research suggests scientists have overestimated capacity of wind farms to generate power (eurekalert.org)
- Protesters say turbines will disrupt migratory patterns of swans near Grand Bend (lfpress.com)
Derk: NOBODY builds a nuke where there’s little demand, Greg. Nobody.
It’s not “where” it’s “when”. Try to keep up.
Derk: “Show me one piece of evidence that your alleged “engineers” have carefully chosen to design a combination of solar and wind to achieve a stable output.”
Well, may be they just got lucky and it designed itself.
Kim 2000 your second link set off my work firewall under the category of “Pornography”. Thanks.
Anthony, there are a few other market ‘tricks’ underway in this new strategy. I have a piece ready to be published at another place and will post a link to it in tips and notes upon publication.
john from DB
For a somewhat cynical view of “wind power”, one might read the novel “Cold Wind” by CJ Box who writes novels about a Wyoming game warden who gets involved in all sorts of nefarious resource schemes. Of course, his view is that “Wind Power” has little to do with producing electricity.
Now, I think that even though advocates of wind power are mostly technical illiterates who go on “feelings” more than reason, we can manipulate their feelings for the common good. We promise to double wind power BUT, because it is unreliable, we have to keep the old coal and old nuke plants online although operating at lower capacity but capable of quickly coming up to produce what the wind is NOT producing. They get the warm fuzzies of having wind power and we get a reliable power grid with coal and nukes usign existing plants.
The criticisms of wind power almost all assume direct connection to the grid. So does the paper above. Why is direct connection (as opposed to buffered connection) a necessary assumption to this debate? Yes, a buffered connection will be somewhat less efficient. arthur4563 estimates a 30% penalty above. So what? How to accomplish it and how much that value can be improved are engineering questions, not challenges to the underlying science (or politics).
Several commentors above have already proposed a few of the buffering options. Flywheels (theoretically feasible though not be my pick for safety reasons), chemical batteries (heavy metals), a superconducting distribution grid (ideal but so far more science fiction than technological fact) and water. That last deserves more attention and perhaps even some deference from the list here. It is, after all, an established practice. See, for example, Kinzua Dam and the Seneca Pumped Storage Generating Station. (The local Park Service claims that they were the first such facility though they are by no means the largest.) The solution is not (merely) to regulate supply but also to regulate demand by using your own buffering operations as part of the consumption pattern.
The approach has some advantages. It takes a lot less time to turn on a pump than to spin up a standby generator. With the right pumps, well, you might not be able to consume unlimited “excess generation” but you could consume an awful lot of it and have very fine control over how much you need to consume at a time. And there is no requirement to co-locate the windfarm and pumped storage facilities – they just have to be within reasonable connection to your grid.
Or maybe there’s a better buffering mechanism. My point is that we should not be skeptical to the point of becoming reflexively contrarian. The proper role of a skeptic is to give your opposition the benefit of doubt and then show that they are still wrong. Criticisms of wind power should assume the best power-buffering system we can reasonably design and show why it’s not cost-effective even then (or, more accurately, to show under what conditions it can be cost-effective and when it can’t). Simple criticisms that unnecessarily assume unbuffered connection to the grid don’t advance the debate.
I have a different idea that actually takes advantage of fluctuating supply. First, imagine your home is completely “off grid” and is powered by a battery pack with sufficient power to provide 24 hours of an average day’s use. Now lets say that battery pack is connected via a smart charging controller to the grid. Lets also say that the grid transmits the current cost of power at frequent intervals, lets say every 60 seconds. When power is more plentiful, the cost is lower, when power consumption rises and begins to reach capacity, the cost increases.
What this does is de-couples your time of use of energy from the time you consume it from the grid and allows you to buy power when it is cheap and use it when it is more expensive. Let’s say you want to shift all of your grid consumption to the late night hours unless the price drops below some amount. If a surge of wind power comes along, the utility can lower the announced price of electricity and the individual homes will increase consumption automatically. If we have a day when power consumption is high and nearing the limits of grid capacity, the price goes up and these homes disconnect themselves from the grid and use the power they bought the night before. Maybe if the price is VERY low (lower than the average cost of the power you currently have stored) you run your house completely off the grid even if the battery is charged fully (late at night AND a lot of wind available). This would be particularly effective for offices that operate 24×7.
If there should be a complete grid outage, one would have power stored to make it through a brief loss without even noticing and in areas where natural gas is available, maybe a natural gas generator kicks in to charge your batteries if they get drained very low.
None of what I wrote above requires “new” technology. I would use AGM batteries. They are very safe, don’t spill, don’t leak, etc.
THOSE CHANGES (in demand) are:
1) much more gradual than ‘waves of wind’ can and will be (consider frontal passage for instance, or ‘outflow’ from a T-storm miles away even, remember, too, in an active power system DEMAND and GENERATION must match or frequency and voltage move around to the point of possibly becoming UNSTABLE)
2) that demand is statistically PREDICTABLE over the course of a day.
For instance, on this chart on the ERCOT (state of Texas’ ISO) webpage note how the *actual* demand consumption pretty well follows the *predicted* demand consumption:
http://www.ercot.com/
One can PLAN FOR the amount of generation to bring ON LINE during the course of a day as demand grows, with wind it can be a whole ‘nother ball game.
.
I simply love the physics behind pepetual wind power generation which will work as soon as the ether supporting the radio waves clears a bit from all that CO2.
Many posters are suggesting pumped storage to ‘smooth-out” wind power production. Well, it is being done, but can only be done for a small portion because the volume is too small. If you look at UK power production sites (there was a great link here a few days ago that I have lost – can someone re-post it?), it showed that there is just 1GW available.
I remember a paper reviewed here a couple of years ago by a physics prof in the UK (no skeptic, if I remember correctly) who calculated that to reach the UK gov’s wind power estimate would mean thousands of pumped storage facilities, covering almost every hill in the UK. I think he even pointed out that the concrete needed to make these facilities would release more CO2 than the wind power would save for quite a few years as well. Once again, the “solution” turns out to have a higher cost than the “problem”.
“American Institute of Physics’ Journal of Renewable and Sustainable Energy,”
That pretty much says it all. Engineers they are not. The name of the journal is further proof of the underlying reason some prominent physicists quit the organization. What is a once prestigious organization like this doing mixing with no-nothing green world-order changers. Give a power engineer the (virtual) “idea” and he will tell you whether you’ve got something or not, Making windmills work better to match the grid is to cut their output even lower than it is now. There is no way to make it greater and certainly no way to make it cheaper and even no way to make it “sustainable” whatever doctrinaire definition of this you have in mind. If you want sustainability, go for a modern designed nuclear plant. There have been remarkably few accidents, even though most of them were designed half a century ago. A modern one would be much lower risk. Or what could be more sustainable than clean coal or natural gas sources – your cooked up CO2 problem is about to take a terminal dive. Rant over.
So just what the hell did they just say was new about their new strategy. So they will generate power when the wind blows, and if it doesn’t, then they won’t; and what’s with this adjust the speed thing ??
Lemme guess, this is some fuzzy logic thing right ?, all assembled out of “maybe” gates.
Do these genii realize how much simpler it is to just turn up the gas, or send some more water flowing through the turbine, when you want more electricity.
http://transmission.bpa.gov/business/operations/wind/baltwg.aspx
You will see that the total generation is always greater than the total BPA load because most of the time BPA is a net exporter of energy. The BPA Load does not include scheduled energy to other balancing authority areas.
That is another solution to load / demand matching: export to neighbouring authorities.
France exports nuke to Germany at night while remaining an net importer of electricity.
These issue are not new and are not restricted to renewables.
A quick update – there are two pumped storage facilities in the UK, one commissioned in 1966 in Scotland with a max power of 440 MW and one in north Wales with a max power of 1,770 MW and that was commissioned in 1983 (I remember a school trip there as youngster while it was being built). The fact that no more have been built during the wind farm era suggests either that they are not all that efficient or useful – or that someone made a big boo-boo!
Need a way to store the energy… best way (if the wind mills are placed on a hill… is the pump storage system.. which would cost more than the windmills….
Heres a big one…
http://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Station
Put a baby thorium reactor at each site. Cant thorium can be ramped up and down or turned off quickly unlike plutonium?
_Jim says March 25, 2013 at 2:45 pm
“THOSE CHANGES (in demand) are:
1) much more gradual than ‘waves of wind’…”
That will depend on the spatial extent and number of wind turbines in a grid.
“2) that demand is statistically PREDICTABLE over the course of a day.”
Weather can be forecast.
The point is that the devil is in the details. It might work in some places and not in others. There is a tendency by some, however, to dwell on the situations where it may be problematic, or assume all problems are universal and insurmountable, while ignoring other evidence.
The worst instance of this problem of which I know is the Pacific NW, where we have to spill water from the giant Columbia River hydrodams to accommodate the wind power surges. The windmills require massive taxpayer subsidies, kill birds & bats that eat the insects which eat our wheat, need massive amounts of CO2 productive concrete, coal-fired plant backups & cause us to lose real renewable power through the dams. A program only socialists could love.
Greg, thanks for this:
Greg says:
March 25, 2013 at 2:15 pm
Well the assumptions that need to be looked at are why would you need to produce a flat rate output 24/7 when demand is not 24/7 constant?
That’s understood, however although demand is (fairly) predictable, supply isn’t.
You might like to take a look at the data raw. The CSV downloadable from gridwatch tracks demand and supply from all sources. I was only interested in wind generation, but there is also data for pumped storage, hydro, coal, imported etc.
Gridwatch: http://www.gridwatch.templar.co.uk/
The other thing, as in my earlier comments, it that solar is generally inverse of wind. Wind/sun ratio is considerably higher for countries like UK of course. I like sun more than wind, so I left.
I’m not sure that’s entirely true. We get very windy sunny days, and very calm overcast or rainy days. And in the UK we don’t really get all that much sun, which is, I suspect, why solar hasn’t really caught on over here.
There are massive hydraulic storage systems in the Scottish highlands that were build to store overnight overproduction from nukes. Same principal can be used to store other energy sources that are not matched to demand.
Wind generation fluctuates across a wide range (e.g. low of 6Mw to high of 5Gw) at all times of the day, and those low periods can last for days. Pumped storage will only fill in for so long before it’s exhausted.
Having said that, pumped could mitigate some of the problem, assuming we can find enough sites; the trouble with such installations is that they tend to be only really practical in mountainous areas, most of which are areas of outstanding natural beauty and planning would (thankfully!) be a nightmare. Worth seeing if I can’t factor pumped storage in somehow though, so thanks for that.
Greg says:
March 25, 2013 at 2:58 pm
“These issue are not new and are not restricted to renewables.”
I beg to differ. As the output of renewables is arbitrary and by law not regulated (When the grid is close to meltdown the authorities BEG wind turbine owners to stop producing here in Germany; there is no law provision that could force them), you simply cannot apply whatever applies to a conventional power source to wind and solar. Therefore, entirely different issues.
I would like to see a Greenie inside every windmill. On a bicycle. Pedalling. Would that do it?
Mike Rossander says:
March 25, 2013 at 2:43 pm
“Several commentors above have already proposed a few of the buffering options. Flywheels (theoretically feasible though not be my pick for safety reasons), chemical batteries (heavy metals), a superconducting distribution grid (ideal but so far more science fiction than technological fact) and water. That last deserves more attention and perhaps even some deference from the list here. It is, after all, an established practice. See, for example, Kinzua Dam and the Seneca Pumped S
torage Generating Station. ”
More than you really wanted to know about how many pumped storage reservoirs a tiny nation like the UK would need to build to make it through a blocking high:
http://www.withouthotair.com/
Written by a warmist climate change advisor to the UK govt.
Spoiler: A LOT.
DBO says:
March 25, 2013 at 2:40 pm
“We promise to double wind power BUT, because it is unreliable, we have to keep the old coal and old nuke plants online although operating at lower capacity but capable of quickly coming up to produce what the wind is NOT producing. They get the warm fuzzies of having wind power and we get a reliable power grid with coal and nukes usign existing plants.
”
Nice plan. It is about what happens in Germany. It has the consequence that the peaker plants are needed only for very few operating hours. This renders them uneconomic, as they used to make their profit by selling during such peak hours for high marginal prices. As they are needed less and less – but during the few times they are needed they are VERY MUCH needed – they can no longer recoup their capital costs. EON considers shutting several gas peaker plants in Germany down.
So as the energy market is totally rigged through the guarantee of always accepting arbitrarily high amounts of renewable energy, for fixed prices, more and more rigging is necessary to stem the tide of unintended consequences, and the system becomes more and more expensive. We’ll surely end up subsidizing the peaker plants as well to prevent the owners from shutting them down – and so you have all that power plant infrastructure sitting there waiting until it is needed.
Environmentalism was always about consuming less yet forces us to double and triple the infrastructure, meaning more concrete, more electronics, more copper, more steel…
This is the result of listening to idiots.
Derek
Way up thread you asked someone to check your calculations that providing the power you suggested would take up more land tan the UK has.
I think you will find the answer in the link above ‘without hot air’.; It was written by Prof Mackay who is climate change adviser to DECC. I think he likened the land needed to Wales, but I do not know the power he suggested would be generated.
tonyb
The historical fact that sails were abandoned for even the very poor early steam engines for marine transport tells you all you need to know about wind power. Works fine on sailboats for fun, nothing serious.
“including fully stopping turbines – in order to avoid any possible damage to the power grid from spikes in supply.”
Forgive them for they know not what they do! With the darn things at best cranking out 10-15% of capacity and delivering less than 1 percent of grid demand you aren’t going to affect a sizeable spike in the grid. The grid already has to deal with this and even knows when it will occur. Interconnection with an unreliable source is a 24/7 headache for the grid. Anyway, spikes and valleys in demand are pretty much related to the clock – wind, not so much. I believe the only way to maximize output from windmills is to be able to store it in electrical storage – pumping water is too big a deal for an effective adjunct to a wind farm, unless you are pumping water back in behind an existing hydro dam in the vicinity.