September 15th, 2012

A statistic you probably don’t know is SAIDI, system average interruption duration index. It means the average number of minutes per year for which the customers of an electricity network are deprived of power, not counting short interruptions of 1-5 minutes, planned interruptions, or “extreme weather events”. It’s one of the two main internationally agreed indicators for grid reliability. Its sibling SAIFI is an indicator of power cut frequency, and is less significant for most customers.

What is it in the USA?
Surprise, surprise, US energy reliability statistics are an uncoordinated mess. The most recent analysis I could find dates to 2008 and the DoE does not (according to Google) publish national SAIDI or other reliability indicators. The latest number I could find is an independent estimate of 240 for 2007. That’s four hours of power cuts a year before the utilities can blame them on ice storms, hurricanes or terrorists.

I thought my Spanish supply was unreliable but the latest SAIDI for Spain is 89. A graph of European SAIDI trends from an authoritative source, the Council of European Energy Regulators (5th benchmarking report, 2011, Figure 2.1. page 27. Data table page 115.) Sorry for the poor screenshot resolution.

So raising the reliability of US electricity to the Greek level of 134 would be a considerable improvement.
Average and best European practice is far below that. The reliability leader is not surprisingly Germany, at 15.3 minutes in 2011.

Notice the way that the arrival of lots of intermittent wind and solar energy in Denmark, Germany, and Spain has been making their grids less reliable? Me neither. The two seem entirely uncorrelated. Intermittent doesn’t mean unpredictable, and competent grid managers have found ways of coping.

It might be an idea for American journalists to actually look at Germany instead of writing stuff like this, from James Conca in Forbes:

The grid can’t handle it [the German renewable energy plan], the transmission system is not there, and the power disruptions and brownouts are wreaking havoc on the country’s energy reliability.

Craig Morris at the German-based Renewables International blog (h/t for my topic) has fun with this nonsense.

The better performance of Europe in general over the USA, and the general trend to improvement, are satisfactorily explained by European coordination and regulatory pressure. It’s still an interesting question why German electricity is at the top. It isn’t formal centralisation. In contrast to the national grid monopolies of Britain, France, and Spain, Germany counts four regional high-voltage grid operators (map here) and a staggering total of 864 network operators.

It would be worthwhile for US policymakers and regulators to find out how this federalised patchwork – rather like their own – works. Non-exclusive hypotheses: the regulator is tough and uncaptured; German technicians and engineers are exceptionally well-trained and competent; operators buy high-quality German equipment; the four large regional networks – the smallest covers a population of 10 million Baden-Württemburgers – impose high standards on the many tadpoles; German consumers have a strong preference for reliability and are willing to pay for it – the average retail price is 23c per kwh); perhaps the law provides for effective compensation for cuts.

A quicker route to progress in the US would be for Steven Chu to publish a reliability league table for electricity, by state and by utility. The shaming effect at the bottom would I think be pretty strong; and a race might also start at the top, once the lag with Europe becomes well known.

I take the opportunity of reposting my favourite pylon snap of am expensive but stunning French design, here on a 400kv line crossing the Somme. SFIK Germany lags in grid aesthetics.

7 Responses to “An ambitious target: Greek power cuts”

  1. dave schutz says:

    Maybe one thing that’s going right for Greek reliability, is no one can afford to buy the power, so there are no demand surges. If you built your system for the predicted demand today based on trends of five years ago, you would have more capacity than you needed now!

  2. Katja says:

    Dear god, this Forbes article is a poorly researched mess. Was it originally written for the WSJ op-ed page, I wonder?

    One thing that the Renewables International post does not even fully address is the “fixed energy credits” thing, which, as far as I know, is made up from whole cloth. If you’re the recipient of Hartz IV benefits (i.e., you’re poor or long term unemployed), you’re entitled to having all reasonable usage of utilities paid for (i.e. relative to typical utility prices and assuming no excessive usage, so if energy prices go up, so does your allowance). This is a hard and inflexible constitutional requirement, a combination of the “human dignity” requirement in article 1 of the Basic Law and the “social state” mandate in article 20; as a consequence, every German is entitled to a life in human dignity, no matter his or her personal circumstances. At a very minimum, human dignity includes not freezing to death in winter, having water available for personal hygiene and cooking, and electricity for light, cooking, etc.

    I’m not sure where he’s getting this “fixed energy credits” idea from, except maybe as an extrapolation from the sorry excuse for a social safety net that we have in America?

  3. Jay C says:

    So the average US power customer has to go without power (outside the given “exceptions”) for 20 minutes every month: i.e. 5 minutes each week? Vs. 1m42s for the average Spaniard, and 29 seconds for the average German? I’m wondering how much this statistic (as are so many American comparisons) is skewed by the relative size, population/population density and grid development differences between the US and most European countries? There are likely to be many more places in the US (especially in the West) where the electric grid is very widespread and serving fewer people (North Dakota, frex) than even in Spain. Which less-reliable grids are likely to pull the US SAIFI average up.

    That said, the difference in attitudes vis-a-vis regulation, renewal energy and technical expertise is striking…

    • Sweden, large, sparsely populated and very cold in winter, comes in at a SAIDI of 89: a lot worse than Germany, but much better than Greece or the US. The first American report I linked to from 2008 (here, page 11) doesn’t suggest that the unreliability is concentrated in the empty West. In fact their “Mountain” region does best at 118; the worst is the quite densely populated Great Lakes region at 498. The explanation isn’t geography.

  4. ferd says:

    I’d LOVE to see U.S. grid reliability stats published loud and clear. I’d also like to see such stats on the companies who build our roads. Who built the road that fell apart in a year? Who did the repair patches that are smooth as silk, and who did the ones that break your shocks. etc.

Post a Comment