Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.
Display:
(well, first year calculus for arts students)

but I spend a certain amount of time looking for trends in data.

So I had a quick look at the raw data (the source Danish spreadsheet with numbers for individual turbines) and yeah, if you graph turbines over 15 years, you can see a clear decline trend.

Depending on the data you choose to look at, you can get something approaching the 20% (i.e. decline from 22% to 18% load factor) cited by ref.org.

Then I thought about the turbines which have 15-year time series. Sure enough, all of them are 600kw or less. The smaller (i.e. older) ones have higher decline rates, probably over 20% over 15 years. The larger machines have smaller decline rates.

Then of course, if you look at the classes of machines which don't yet have 15 years of service, the picture changes...

No surprises here. "Lying with data 101" covers it.

It is rightly acknowledged that people of faith have no monopoly of virtue - Queen Elizabeth II

by eurogreen on Mon Dec 24th, 2012 at 08:43:58 AM EST
[ Parent ]
If this is what no mathematical training looks like, I'm more challenged than I thought.

You can't be me, I'm taken
by Sven Triloqvist on Mon Dec 24th, 2012 at 08:52:12 AM EST
[ Parent ]
I can find no basis at all for the alleged steep decline in Danish offshore wind capacity factor :

The rate of decline in performance is greatest for offshore installations in Denmark, with a fall from load factors of over 40% at ages 0 and 1 to less than 15% by at ages 9 and 10.

Using all the available data in the raw data source cited (I note that there are only two installations with more than ten years of history!), I get the following graph when I calculate load factors :


So let's see : those installations that had over 40% load factor at age 0 still have over 40%, however none of them have yet reached age 9 or 10. What's more, no Danish offshore installation has a load factor of less than 15%. The two oldest are at 20% and 32% respectively in 2011.

Come on Jake, can you find a way of excusing, or even explaining, the way he gets his takeaway quote?

It is rightly acknowledged that people of faith have no monopoly of virtue - Queen Elizabeth II

by eurogreen on Mon Dec 24th, 2012 at 10:03:20 AM EST
[ Parent ]
That's odd.

If he actually implemented the protocol he states he is implementing on that data, he should not get the results he does. He claims to be doing perfectly ordinary within estimator on fixed effects panel data modeling. That shouldn't turn a figure like the one you have into a figure like the one he gets.

Unless of course he tortured his model in a way that he should know (because it would trigger more than one of the several misspecification tests built into all modern econometric software packages) is wrong.

Good catch.

- Jake

Friends come and go. Enemies accumulate.

by JakeS (JangoSierra 'at' gmail 'dot' com) on Mon Dec 24th, 2012 at 11:44:12 AM EST
[ Parent ]
Most of these are small. Nysted (2003) and Horns Rev (2002) should have the biggest weighting, but while I recognise Horn Rev's name in the list, I don't see Nysted. Both, as far as I know, have good performance today (and, as I mentioned before, Nysted was sold recently 50% to a pension fund so they would have not done that if performance was declining...)

Wind power
by Jerome a Paris (etg@eurotrib.com) on Mon Dec 24th, 2012 at 12:13:16 PM EST
[ Parent ]
A word on "methodology", i.e. wot I done.

From the source data (Excel file AnlaegProdTilNettet.xls alleged to come from the Danish energy web site), I filtered the offshore projects. The data is presented as production figures from individual turbines, but in reality it's clearly some sort of composite figure, as each machine of a given type in a given park has the same numbers. So I chose one line from each group. I've double checked, and in the source data, the projects aren't referenced by their habitual names, but the correspondances are clear with a bit of digging.

In the above graph :

  • Uoplyst = Vindeby
  • Ukendt = Tunø Knob
  • 2nd and 3rd Uoplyst = Middelgrunden
  • Horns Rev = Horns Rev
  • 2nd and 3rd Ukendt = Rønland 1
  • Hav = Samso
  • The last Uoplyst = Nysted

So this graph contains no aggregation or averaging of any kind, other than that of the source data, and represents capacity factor by year of all the pre-2009 Danish offshore parks.

I eat raw data for breakfast. Great bleeding chunks of it. But I'm a moralist, and I don't like to see data tortured. Clearly the above data must have suffered a great deal to get it to "confess" the trend that Hughes gets from it. Because the trend just isn't there.

There are too many British onshore projects to submit them to the same naive treatment, but from looking at the data, the story is the same.

It would be worth keeping an eye open for press take-up of this atrocity, in order to "set them straight". Better would be to pre-empt them of course...

A quick google shows that there hasn't been much take-up yet, though the FT has published an article, which I am composing a reply to.

It is rightly acknowledged that people of faith have no monopoly of virtue - Queen Elizabeth II

by eurogreen on Wed Dec 26th, 2012 at 04:36:56 AM EST
[ Parent ]


It is rightly acknowledged that people of faith have no monopoly of virtue - Queen Elizabeth II
by eurogreen on Wed Dec 26th, 2012 at 04:46:54 AM EST
[ Parent ]
You see clearly the years when the big farms had to deal with serial defects  - very early on at Horns Rev (where all the gearboxes had to be replaced within the first year or two), and after a few years at Nysted (where the gearbox problems were less publicized but did require large scale action with the same performance impact as on Horns Rev). Anf both are now performing at high levels (and have given their respective manufacturers, Vestas and Siemens, massive experience which they are now using profitably on their new turbine models)

Wind power
by Jerome a Paris (etg@eurotrib.com) on Wed Dec 26th, 2012 at 10:51:38 AM EST
[ Parent ]
Notice also the strong inter-annual variation, which is normal and independent of machine performance.

Offshore wind technology is moving completely away from the standard technology represented by these statistics.

"Life shrinks or expands in proportion to one's courage." - Ana´s Nin

by Crazy Horse on Wed Dec 26th, 2012 at 12:44:16 PM EST
[ Parent ]
And Vindby and Tunø Knob are onshore projects which happen to be in water at the coast, though they were called pioneering offshore. Of course they have the best capacity factors possible onshore, but don't reach current offshore standards. They are meters offshore, if my memory hasn't rusted.

"Life shrinks or expands in proportion to one's courage." - Ana´s Nin
by Crazy Horse on Wed Dec 26th, 2012 at 12:49:36 PM EST
[ Parent ]
If you check the report, he actually shows himself the lack of a clear trend in raw data for UK on-shore, too (page 24):



*Lunatic*, n.
One whose delusions are out of fashion.

by DoDo on Sat Dec 29th, 2012 at 06:58:47 AM EST
[ Parent ]
Yep. Teletubbie version: newer wind farms performed well and continue to do so; older ones performed more poorly and still do. Ergo: performance diminishes with age... hmm, isn't this what they call the 'ecological fallacy'? The confounder being the year of construction.

Why am I reminded of the patricide asking the court for clemency on the grounds of being an orphan?

by mustakissa on Tue Jan 1st, 2013 at 09:33:46 AM EST
[ Parent ]

newer wind farms performed well and continue to do so; older ones performed more poorly and still do. Ergo: performance diminishes with age... hmm, isn't this what they call the 'ecological fallacy'? The confounder being the year of construction.

Makes a lot of sense...thanks.

Wind power

by Jerome a Paris (etg@eurotrib.com) on Wed Jan 2nd, 2013 at 12:06:22 PM EST
[ Parent ]
What you have pointed at is standard when a new technology is introduced into the market.  The IBM 1401 solid state computer was a huge hit in the market, solving a number of major business problems, when it was introduced in 1959.  Five years later IBM introduced the System/360 a technically superior product to the 1401, yet the 1401 continued to sell until it was withdrawn in 1971.  Even after IBM no longer sold the 1401 the machines continued to adequately accomplish business tasks; I worked on a 1401 in 1976 (IIRC) programming new business tasks and extending functionality of existing programs.

Let me list a couple of things:

  1.  People invest in technology when the existing functionality meets their needs (or financial forecasting, financial ratios, cash flow analysis, & etc.) at a specific point in time.  

  2.  The GO/NO-GO decision to purchase technology at a point in time is made with the knowledge at that time.  (Like, duh.)  

  3.  It is doubtful the "superior" IBM System/360 would have been built if the "inferior" 1401 had failed in the market.

  4.  It doesn't matter if a "superior" machine is available latter when the "inferior" machine is meeting the needs, tasks, goals, financial return, etc. assigned at the time of the GO decision.

  5.  Both users and producers of technology climb a learning curve.  The users learn how to use the stuff better, the producers learn how to make the stuff better, and (ideally) they interact in a virtuous positive feedback to the benefit of both.


She believed in nothing; only her skepticism kept her from being an atheist. -- Jean-Paul Sartre
by ATinNM on Wed Jan 2nd, 2013 at 12:41:51 PM EST
[ Parent ]
Yep. And once a wind park has been built, the electricity it produces is essentially (almost) free. No reason to take it down unless you're short of sites for new ones.
by mustakissa on Wed Jan 2nd, 2013 at 02:31:56 PM EST
[ Parent ]
Some of the oldest commercial WTs in the world were allowed to run up to 30 years before repowering began, with many reaching 25 years.

They were known as "cash cows."

"Life shrinks or expands in proportion to one's courage." - Ana´s Nin

by Crazy Horse on Wed Jan 2nd, 2013 at 02:49:48 PM EST
[ Parent ]
I could write a book on the stupidity of using information known at Time Stamp + 20 to poo-poo the return of tech installed at Time Stamp + 0.

She believed in nothing; only her skepticism kept her from being an atheist. -- Jean-Paul Sartre
by ATinNM on Wed Jan 2nd, 2013 at 03:00:05 PM EST
[ Parent ]
A couple of past ET references to the Renewable Energy Foundation:

One

Two

by afew (afew(a in a circle)eurotrib_dot_com) on Mon Dec 24th, 2012 at 11:53:17 AM EST
[ Parent ]
One more note (received by email):


"Professor", Hughes did some earlier work on how expensive the gas backup to wind was, which was published by that former Tory Chancellor's so called think tank. (I refuse to clutter up my mind with their names!)
It was extensively and conclusively debunked by Robert Gross of Imperial, who gave evidence to the relevant Parliamentary committee.
I commented on the FT. http://www.ft.com/cms/s/0/2e0945e2-4a09-11e2-8002-00144feab49a.html#comment-3447592


Wind power
by Jerome a Paris (etg@eurotrib.com) on Mon Dec 24th, 2012 at 12:11:11 PM EST
[ Parent ]
That would be Nigel Lawson's Global Warming Policy Foundation. As deceptively named as the Renewable Energy Foundation. For Lawson's tink-tonk's links to a certain type of energy industry, see here.

For Hughes and Gross, see here on ET.

by afew (afew(a in a circle)eurotrib_dot_com) on Tue Dec 25th, 2012 at 03:08:23 AM EST
[ Parent ]
So we knew of him?

Then the analysis could have been shorter, as one could have disbandonded assumption of good faith and centered on where lies can be hidden.

Sweden's finest (and perhaps only) collaborative, leftist e-newspaper Synapze.se

by A swedish kind of death on Tue Dec 25th, 2012 at 09:09:42 AM EST
[ Parent ]

Display:

Top Diaries

Occasional Series