Brits to shut grid for solar flare

[Allan_L (Allan)]

Just saw this today.
Thought it may be worthy of publication here.

Originally Posted by Publication: The Australian (p6, Mon 20 Jun 2011)

Britain is preparing to invoke emergency powers to turn off its electricity to shield the country from the worst effects of the biggest solar flare in 150 years. The British government has been warned that a massive surge of energy from the sun could hit the Earth in the next 18 months. In a worst-case scenario, it could blow out the national grid and leave parts of the country without electricity for months.

[supernova1965 (Warren)]

I would rather this than months of outage I would hope all countries would take the same measure seems sensible to me

[higginsdj]

OMG - and who are they allegedly getting their advice from? I guess someones got caught up in the 2012 hysteria!

Cheers

[supernova1965 (Warren)]

Originally Posted by higginsdj

OMG - and who are they allegedly getting their advice from? I guess someones got caught up in the 2012 hysteria!

Cheers

But we would have enough time for an emergency shut down once a flare was sighted if we had an eye on the sun I believe it takes 8.3 hours for the flare to arrive would it not be a wise move to shut down the grid for an hour or so

[Barrykgerdes]

There must have been hudreds of flares in the last 100 years. Is they any record of effects other than long range radio communication problems.

The warning seems to be just an excuse to shut down power when some politition wants to show his power.

Barry

[gary]

Originally Posted by Barrykgerdes

There must have been hudreds of flares in the last 100 years. Is they any record of effects other than long range radio communication problems.

Barry

Hi Barry,

Indeed there is, the most famous probably being the March 13th 1989 event with
Hydro-Quebec Power. The solar storm brought down a sizable part of the Quebec
grid and chaos ensued.

With night time temperatures in Toronto well below zero, millions found themselves
cold and without power. The airports were closed and schools and businesses
were shut as was the metro. Street lights, traffic lights and so on went out. Some people
got caught stuck in dark buildings.

Other grids in New York and New England were also affected by the same solar storm.

This particular incident comes up in the professional electrical engineering
press now and then. Ever since that event, power authorities at those latitudes
seem to have taken it on as a cautionary tale.

Best regards

Gary

[tonybarry (Tony)]

Yes, Barry, have a look for "coronal mass ejections" on google.

They stand a chance of seriously damaging the electrical grid, especially transformers (which are not easy to replace).

Regards,
Tony Barry

[gary]

Some readers might find this six page November 2000 article from the Institute
of Electrical and Electronic Engineers (IEEE) Spectrum magazine
entitled "Shielding grids from solar storms" of interest -
http://thayer.dartmouth.edu/spacesci...molinski00.pdf

Written 11 years after the Quebec event, the authors highlight some of the ways
power grids are susceptible to "geomagnetically induced currents" (GICs) caused
by solar activity and discuss some of the remedial steps power utilities can take.

Best regards

Gary Kopff
Proud member IEEE 33 years

[CraigS]

Solar storms aren't the only thing which has taken down North East Canada's power grid, either .. the North American ice storm of 1998 took out 3 million people's power for up to a month and a half.

Their grid still remains at risk from similar events. The UK, Arkansas and Moscow have also had similar vulnerabilities exposed recently.

I wonder how the risks compare (which of the two is more likely), and which of the two has a greater sustained impact ?
(It kind of puts it all in a clearer perspective, looking at it this way).

Cheers

[gary]

Originally Posted by CraigS

I wonder how the risks compare (which of the two is more likely), and which of the two has a greater sustained impact ?

Hi Craig,

You might be interested in pages 2 to 3 (article pp 56-57) in the above cited IEEE Spectrum
article in the section under the heading "Risk Factors".
(Article here - http://thayer.dartmouth.edu/spacesci...molinski00.pdf )

The authors point out that -

Originally Posted by IEEE Spectrum Nov 2000 page 56

"the danger of power systems suffering catastrophic events (outages or damages)
from GICs is modest compared to other hazards.
Most utilities in North America design their physical infrastructures to withstand
any wind storms or ice loadings except for a worse-case event - expected to happen
no more than once every 50 years. The risk of catastrophic damage or an outage
in any given grid system from GICs probably is much smaller than that. What is more,
damage incurred from a severe terrestrial storm would generally be more costly and
difficult to repair.

[gary]

Two articles published in the UK's Independent that discuss where the
US Space Weather Prediction Center are collaborating with the UK Met Office
to "improve solar weather forecasting to a point to warn power companies of
of an imminent storm" arguably highlights the difference between what was
actually said by the original spokesperson and what the press like to report.

Originally Posted by Tom Brogden US Space Weather Prediction Center and quoted in the Independent

There is a feeling that if a "category 5" solar storm – the biggest of the five categories – were to be predicted, then taking the grid off-line before it is due to hit Earth and letting the storm pass would be better than trying to keep things running, he said.

In 1989, a solar geomagentic storm knocked out the electricity grid across large parts of Canada. The loss cascaded across the United States and caused power problems as far away as California. The greatest fear is a massive storm as big as the one documented by astronomer Richard Carrington in 1859, which burnt out telegraph wires.

"The sort of storms capable of doing that are fairly rare events. We refer to them as 'black swans'," Dr Bogdan said. "If the Carrington event occurred today, and power grid operators did not take efforts to safeguard their infrastructure, then we could be facing a scenario like that."

See http://www.independent.co.uk/news/pe...g-2296759.html
http://www.independent.co.uk/news/sc...d-2296748.html

[oosh (Jason)]

What I'd really like to know is if other countries, namely ours, have any procedures in place incase CME's do get thrown at us - I've only heard about the UK in the past. (though that said, I haven't been actively searching for information either)

[CraigS]

Interesting read, Gary.

Just checking on the frequency of occurrence of power effecting ice-storms .. Wiki lists the following:

- 1998: 3 million people w/o power for up to 1.5 months;
- 2008 NE USA: 1.25 -1.3 million people w/o power for 4 to 16 days;
- 2009 USA: 2 million w/o power, killed 55 people;
- Xmas 2010: Moscow blackout of airport and city connecting railway followed by blackout of two other airports causing a complete air transport collapse.

I notice the IEEE article was written in 2000, (mostly) before the above 2008-2010 chaos occurred. Perhaps they've now updated their one-in-fifty year frequency-of-occurrence estimates.

Anyway, how accurate are GIC forecasts ?

I remember lots of speculation about how long the solar minimum was going to last, just recently. It was a big 'surprise' to all the forecasters that it was 'delayed', and went longer than they anticipated.

Methinks solar storm forecasting is still in its very, very, early stages. We seem to expect so much of weather forecasting technologies thesedays … completely forgetting the unpredictability of Chaotically behaving phenomena. The effect of such phenomena is well worthwhile taking defensive measures against, but placing any precise reliance on the forecasts, (ie: like 'the next 18 months'), might be futile and really where the politics of 'perception shaping' kicks in. (My 2 cents worth).

Cheers
PS: Oops .. there was another big outage-causing ice storm in 2005 as well.

[gary]

Originally Posted by CraigS

Interesting read, Gary.

Just checking on the frequency of occurrence of power effecting ice-storms .. Wiki lists the following:

- 1998: 3 million people w/o power for up to 1.5 months;
- 2008 NE USA: 1.25 -1.3 million people w/o power for 4 to 16 days;
- 2009 USA: 2 million w/o power, killed 55 people;
- Xmas 2010: Moscow blackout of airport and city connecting railway followed by blackout of two other airports causing a complete air transport collapse.

I notice the IEEE article was written in 2000, (mostly) before the above 2008-2010 chaos occurred. Perhaps they've now updated their one-in-fifty year frequency-of-occurrence estimates.

Anyway, how accurate are GIC forecasts ?

I remember lots of speculation about how long the solar minimum was going to last, just recently. It was a big 'surprise' to all the forecasters that it was 'delayed', and went longer than they anticipated.

Methinks solar storm forecasting is still in its very, very, early stages. We seem to expect so much of weather forecasting technologies thesedays … completely forgetting the unpredictability of Chaotically behaving phenomena. The effect of such phenomena is well worthwhile taking defensive measures against, but placing any precise reliance on the forecasts is futile and really where the politics of 'perception shaping' kicks in. (My 2 cents worth).

Cheers

Hi Craig,

Thanks for the statistics on the ice storms, which as far as any risk assessment
the power utilities will be making, would clearly rank significantly higher compared
to the risk of a significant GIC event occurring.

With regards the question of how accurate the GIC forecasts are, there was
a paragraph in the Independent article where they quoted Tom Bogdan of the
US Space Weather Prediction Center -

Originally Posted by The Independent

The UK Met Office and Dr Bogdan's centre are now exchanging computer models and expertise in the hope that each can learn from one another about the vagaries of space weather prediction, which Dr Bogdan freely admits is still in its infancy: "Today, unfortunately, space weather is where meteorology was at the end of the 1950s."

The Nov 2000 Spectrum article mentions various possible countermeasures such
as three-phase, three-leg core design transformer designs and one would then
need to multiply the risk of a storm creating significant GICs with the risk
of a given grid with a given architecture failing when presented with such a GIC.

I would guess where Tom Bogdan would be coming from is that space weather
is the profession he is in and given the increasing importance today of infrastructure
such as power grids, satellite communications, GPS and so on, then it would be
prudent to be able to model space weather events more accurately such that
a true risk assessment could be made. Bogdan himself is saying the risk is
very small, but these days one would guess that if it were to occur again like
in Quebec in 1989, it would be an expensive failure.

One of the things the 2000 Spectrum article mentioned was that "The main reason
power systems are increasingly likely to fall victim to GICs is that as electricity
is traded over greater distances, the transmission lines are exposed to larger
induced voltages."

Rather than just wind and ice or solar storms, it seems as if the reliability of a
grid can also be influenced by politics, deregulation, market forces,
market manipulation and so on.

Over the last few years, there have been a few articles in Spectrum about both the
hazards and potential advantages of the trend in the US to deregulate power
and to create ad-hoc national grids where power is traded. Many will recollect the
rolling power cuts in California in 2001 and the Enron scandal.
See http://spectrum.ieee.org/energy/poli...ked-the-grid/1

In a June 1999 Spectrum article, the authors point out that the US electric power system
is the largest and most complex in existence, and raised concerns that with
deregulation, emphasis seemed to shift from "reliability as the mainstay of the
nation's essential power base to reliability as a commodity in the power grid."
See http://spectrum.ieee.org/energy/the-...-the-lights-on
This article predated the Enron scandal surfacing.

[CraigS]

Thanks for all that, Gary.

I think my conclusion would be that there are many more factors which can (and do regularly), result in failure of power grids (other than GIC events).

'Trial runs' or practising turning off the power grid will come at a huge cost, and one would have to carefully balance the true benefit of doing this, against the cost, and a realistic view of the risk of random detrimental events.

Passing legislation seems like a first step. Without emergency legislation, pro-active management is not possible.

Cheers

[midnight (Darrin)]

Loss of power on a grid is quite serious in that you could end up in a "black start" situation whereby most coal fired power stations can not start up by themselves. They need a large (at least 30MW) machine nearby (hydro, GT etc). Australia has some of the best regulations for grid connection and compliance and there is a lot of mathematics behind it - believe me.

The actual cause is induced EMF on the transmission line which can have a large DC component (or VLF - ie < 1Hz) and this finds its way to a transformer which then saturates the core of the transformer. This results in overfluxing and rapid increases in iron core temperature. All transformers work on AC and hence the cycling of the AC voltage prevents core saturation. It is a characteristic of iron.

The phenomenon is real and taken very seriously by transmission authorities. Transformers take many months to build and then transport and installed. If one has overflux protection it may catch it although it is not necessarily designed for this type of influence.

Darrin...

[space oddity]

A large CME hitting the power grid could easily overload the system, blowing transformers and overloading circuits. Given that there are not many spare transformers of the size used in power generation, restoring electricity would take quite some time, especially if the effect was global. In 1859, the Carrington event was almost certainly a CME. If an event of this magnitude were to strike the earth today , with our reliance on all things electronic, we would be in big trouble. As long as they do not get too paranoid, this may well be a wise decision. Remember, another big CME must hit planet earth sometime.

[Allan_L (Allan)]

Oddly enough, this is dealt with at length in the current month's (May/June issue) Australian Sky and Telescope. They at least seem to be treating it as a serious threat.

Whereas, the article I originally quoted went on to say that Australian authorities were not going to worry about it, with a "she'll be right, mate" attitude.

[oosh (Jason)]

Originally Posted by space oddity

A large CME hitting the power grid could easily overload the system, blowing transformers and overloading circuits. Given that there are not many spare transformers of the size used in power generation, restoring electricity would take quite some time, especially if the effect was global. In 1859, the Carrington event was almost certainly a CME. If an event of this magnitude were to strike the earth today , with our reliance on all things electronic, we would be in big trouble. As long as they do not get too paranoid, this may well be a wise decision. Remember, another big CME must hit planet earth sometime.

Indeed, I remember reading we're well overdue for a CME (every hundred years approx?) so I think caution is a wise idea!

[higginsdj]

We get many CME's a month. One is due to hit in 30 minutes......

Cheers