Monday, August 24, 2009

How wind power let the BPA down

Balancing Authority Load and Total Wind Generation Chart, Last 7 days
Link here.

22 comments:

Jason Ribeiro said...

Charles, does the BPA publish this on a regular basis somewhere? Do you have a link?

Charles Barton said...

Address:
"http://www.transmission.bpa.gov/Business/Operations/Wind/baltwg.png"

Anonymous said...

Looks like ~ 5 GW of baseload nuclear could provide almost all power of the BPA. With some emergency resistance heaters to dump the relatively small excess in weekends. Looks like solar may be a reasonable contender for ~ 1 GW of daytime peaking on top of that, especially with a few hours of storage (electric cars maybe?)

Wind with a day of storage would also do fine. Electric cars won't be enough for a dominant role of wind though. Cost of grid energy storage is going down but not very fast, hopefully there will be some major breakthroughs soon...

donb said...

If you check out the link to the graph, just above the graph you will see a link to the data used to generate the graph (in tabular form). It show load and wind power generation in 5 minute intervals.

I downloaded that data yesterday just before noon. I pasted it into an Excel spreadsheet, and ran the correlation function between the load data and wind power generation data. It came up with a value of -0.202. If wind generation moved perfectly in step with power demand, the correlation would be +1. If wind generation moved perfectly contrary to demand, the correlation would be -1. If wind generation were random with respect to demand, the correlation would be zero.

The correlation value of -0.202 indicates the wind generation is fairly random with respect to demand. What correlation is does have to demand indicates that it moves AGAINST the demand curve (tending to go away when you most need it).

Duncan said...

Anonymous,

1 GW of solar in the pacific northwest? They'd be better off betting that HVDC lines will be economical enough so they can import some solar from the desert southwest.

Also, Charles' chart does show it but there was a 4 day stretch in their from Aug 17-20 where wind production was at 10%. How would 1 day storage help bridge a 4 day gap?

Not that 4 days is all that much. Looking back through BPA's data, there was a 2 week stretch in January where wind production was like 1% of capacity. There's no storage system that could handle 2 weeks of outage.

Charles Barton said...

Inthe caseof the BPA, base electricity is hydro, with wind as a supplement. This is a situation in which wind works to advantage, and that is an exception.

Anonymous said...

In the case of lots of hydro it's different yes, in fact Charles is exaggerating a bit when he says wind is letting the grid down due to the large buffer capacity of hydro. But without large amounts of hydro, a four day stretch would be covered by longer backups like piled biomass, biogas (natural gas most likely for the immediate future)... the point about storage is to deal with the brunt of variability, not with all. There's plenty of waste biomass streams to get that last 5 - 10 % of total, no point in getting weeks of electrical storage, unless when one day perhaps electrical storage would be incredibly cheap.

It's not an all or nothing equasion. 100% wind and months of storage or 100% nuclear with energy dumping and no storage, these are all extremes that suboptimal and costly.

Charles Barton said...

The useof biomass amounts to mining the soil for nutrients and minerals, which arethen burned rather than returned to the soil. It isaviolation of the principlesof goodsoil management, and therefore is an unsustainable solution. In thecaseof the BPA wind I was pointing to a larger reality. The graph demonstrate two common realities. Some days there is virtually no electrical output from wind. On other day, wind output drops as demand rises, thus wind is too unreliable to be counted on. In its inability to respond to demand, wind lets the BPA down. This is the rule rather than the exception. The exception is the BPA's relative abundance of hydro power.

Soylent said...

I don't believe so.

The most effective way to destroy soil carbon is to use lots of tilling. With effective herbicides and GMO crops it's quite easy to get away with no tilling at all.

Very little of the 'waste biomass' above ground actually becomes soil carbon and sticks around for more than a few years.

If you optimize for producing char with low-temperature pyrolysis you can make some very stable soil ammendment that seems to work as well as humus. You'd still get some interesting industrial raw materials that can be used for plastics and other stuff. I'm not advocating pyrolysis because I have no idea if it makes any financial sense at all.

Anonymous said...

Biogas is also interesting since virtually all of the nutrients will be concentrated in a slurry that can be easily be used as bio fertilizer. And it benefits from a huge natural gas infrastructure. The trick will be to get the capital cost (per kiloWatt electrical installed) of the digester systems down.

donb said...

Charles Barton wrote:
In the case of the BPA, base electricity is hydro, with wind as a supplement. This is a situation in which wind works to advantage, and that is an exception.

True enough, but there have been occasions when wind generation was max'ed out during times of low demand, and it threatened to destabilize the power system. The normal fix to this is to order a reduction in generation. Unfortunately, wind generation is "must take", so the system operators need to scramble to keep things from falling apart.

Anonymous said...

BPA's worst case scenario is a dual unit trip at Palo Verde - a contingency that has occurred in the past. We all know nuclear runs 90% of the time but few realize how a fault at a NPP can cascade through a grid.

Wind is not a "must take" resource. The economic rules prefer wind but when reliability is at stake you can trim the blades or disconnect the generators altogether. That's what real-time operators do and it's as simple as pressing a few buttons.

The projected overnight costs for new NPP are $4 billion per GW or higher. At the moment this is nuclear's one big trick.

At $2 billion per GW wind is expensive too but wind costs look to come down appreciably in the next few years. Purely on the basis of electricity costs I'd bet on wind right now. At some point wind's variability will make it unwieldy but we're not there yet.

Charles Barton said...

Anonymous the claim that the price of wind will soon come down is not born out in observed price trends. The price of windwent up steadily between 2002 and2008, morethan doubling.

The $2 billionper GW figure refers to name pl;ate power, not power output. Typically it takes windmills with 3 GW nameplate power to equal the average output of a 1 GW nuke.

Anonymous said...

Have you seen the latest LBNL report?

http://eetd.lbl.gov/EA/EMS/reports/2008-wind-technologies.pdf

See page 34 for a $/kW trend. 2002/2009 look like $1500/kW and $1950/kW respectively. The low point looks like $1200/kW in 2004.

As far as falling wind costs go the report simply says, "...evidence of softening turbine prices began to emerge in late 2008." Given the state of the economy I have little doubt that wind turbine costs have been dropping.

"Typically it takes windmills with 3 GW nameplate power to equal the average output of a 1 GW nuke."

I agree but matching up average output doesn't tell you what the underlying costs of electricity are.

See page 26 of the LBNL report for a graph of wind prices. The graph represents the sales price of the electricity after the Renewable Energy Certificates (median price of $20/MWh) and the Production Tax credit ($15-$20/MWh) are taken into account. Stripping out the REC and PTC gives you a ballpark 2008 generation cost of 8 cents per kWh. This compares to MIT's ballpark generation cost for new nuclear of 8.4 cents per kWh.

http://web.mit.edu/ceepr/www/publications/workingpapers/2009-004.pdf

Charles Barton said...

your findings on wind cost are similar to mine, with the exception that downturn in current wind cost is due to the economic down turn, and not due to any underlying cost trend. An economic recovery is likely to bring a return in wind inflation. If you are interested in future costs, the EIA's latest 2016 estimate is for a levelized cost of 141.5 for wind and 107.3 for nuclear. These cost estimates do not include do not include the cost of grid expansions necessitated by far from market wind farms, or the cost of energy storage, or maintaining a fossil fuel back up system that equals 100% of wind capacity. Above 20% grid penetration by wind storage and redundant wind capacity come to the fore, and significantly increase wind related expense.

As for the alleged eight cent cost of wind, in fact wind producers are often forced to sell at a discount because the wind blows at the wrong time. Consider the fact that during a large fraction of the time, Texas wind producers are actually forced to pay the grid to take their electricity. Most Danish wind generated electricity sells on the spot market for something like 14% of Danish retail electrical costs. If wind levelized costs run at 141.5 in 2016, selling wind at 8 cents would be an economic disaster for the wind producers.

Anonymous said...

Wind turbine prices have gone up in recent years but technological innovation and reduced overhead costs due to scale up of order sizes have in fact reduced wind turbine cost. This has created an unsustainable market situation because of very high gross profit margins of turbine manufacturers/sellers.

The slightest market change can now lead to a rapid drop in turbine prices; examples are the continuing economic downturn, any drop in subsidy levels per kWh or per kW, or increased pressure from project owners/buyers asking manufacturers/sellers to reduce turbine list prices...

Increased prices can be bad for consumers on the short term. However, from an innovation perspective, a higher gross margin has some advantages because it attracts more players with innovative approaches to wind technology and allows existing companies to invest more in RD&D.

It's an expensive (fairly inefficient) way to induce innovation, though...

Charles Barton said...

Anonymous, your story is that the rise of windmill prices is due to manufacturer greed. But it should be noted that the cost of all energy generations technologies rose during the same time period, and underlying costs, for example the cost of labor and materials rose as well. It is remarkable how people who insist that the price of windmills is going down, ignore all of this.

Anonymous said...

Do you agree with the LBNL prices/costs? LBNL's data does not agree with your statements by a wide margin.

Charles Barton said...

My own samples of wind projects costs during 2007andthe first half of 2008 ran between $2200 and $2800 per kW per installation, based on a sample of 8 projects. I would have to review the LBNL sampling process, but their prices are substantially lower than the ones I verified.

Anonymous said...

The cost of materials has dropped and this will be reflected in later turbine prices, it does take time because of the delay induced by working with longer term (multiple months) contracts...

Materials cost, however, do not nearly explain all of the cost increase. It's also increased margins that explain that.

Greed is not the word I'd use. Fast growth does require higher margins (if you want to stay out of cash flow troubles and keep the investors happy). Also, if fossil fuels cost rise, wind turbine manufacturers are in a better position to charge more for their turbines. Why shouldn't they, it's the market and they're a commercial company. And with feed in tariffs being reduced only slightly, there is less stimulation to reduce cost from a policy push as well. But innovation an economies of scale are still continuing incrementally, destabilizing the price umbrella that we have today. Maybe there will be a shakeout of wind technology and related companies...

I don't know enough about the nuclear industry situation to know what the likely course of reactor prices may be. But the issue of price vs production (real) cost is interesting. Maybe you could do a post about it sometime.

Charles Barton said...

I suspect,although it has not been reported, that orders for turbines have been canceled. However, with the push of the Obama Administration to fund new renewables,and the installation of large numbers of turbines in China,we probably are going to see upwards price pressure on turbines again. I suspect that what you call greed would be called by most economist the normal operation of price in a market where the demand for certain goods outstrips production.

Anonymous said...

Actually Charles you're the one who called it greed if you check the posts.

If demand outstrips production consistently for a longer period, this will attract new players because of increased profitability, unless there are fundamental material constraints (not really the case for wind turbines except for extremely high growth rates). Sooner or later, the price will come down. The wind market also doesn't have to grow exponentially forever, just a decade more and then declining growth rates would make wind very dominant.

Followers

Blog Archive

Some neat videos

Nuclear Advocacy Webring
Ring Owner: Nuclear is Our Future Site: Nuclear is Our Future
Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet
Get Your Free Web Ring
by Bravenet.com
Dr. Joe Bonometti speaking on thorium/LFTR technology at Georgia Tech David LeBlanc on LFTR/MSR technology Robert Hargraves on AIM High