Rules and Guidelines
Want to participate in the discussions? Please read our Community Rules and Guidelines. Need some help? Visit Support and Answers.

  • State Not Answered
  • Locked Locked
  • Replies 113 replies
  • Subscribers 32 subscribers
  • Views 6686 views
  • Users 0 members are here
Options
You may also be interested in
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Electrical outages. cyber attacks ?

aligarjon
What's the chances of the power outages and airport problems being cyber attacks.     Is that possible.   I would think so  ?


Gary

  • 0
    My view is that failures of this nature are acceptable provided that they remain very rare. It is not worth spending vast amounts of money to mitigate something that happens a dozen times a century.

    If however this sort of failure becomes more common, then something needs to be done. This "something" could consist of large flywheels, batteries, or the prompt cutting off of customers who have agreed to this in return for a preferential tariff.
  • 0

    National Grid have released their interim report. I've attached it



    Thanks for hat Jam - it's an interesting read.


    So in short a lightning strike on a distribution line had three effects:

    1. Tripped out the protection for that line (as intended), losing both loads and embedded generation in the area served, then the protection autoreclosed within 20s restoring the loads, but embedded generation has to wait longer before it's allowed to reconnect - so the load on the national grid increased by the amount of the embedded generation (500MW they say).

    2. The spikes from the strike also caused two of the three interconnectors from a wind farm to trip out (not expected) - losing 737MW

    3. The spikes from the strike also caused one steam turbine at Barford Gas Power Station to trip out (not expected) - losing 244MW


    The loss of the steam turbine knocked the combined cycle system out of kilter, so causing the two associated gas turbines to trip out within 90s - losing a further 397MW


    So the grid was then out of balance to the tune of 1.9GW - while the grid has 'reserves' ready for only 1.0GW - enough to compensate for the loss the largest single generator - so the DNO's systems started load shedding.


    (Is it just me or does 500MW of embedded generation on one single distribution line sound like a lot - especially at that time of day when solar would have been very much in decline? I wonder if Mike's point about modern electronic (VF drives etc) loads increasing demand in response to low voltage - unlike old fashioned star/delta induction motors that would probably act as mini generators for a while - might have accounted for some of that figure)


    Maybe we just need a few more SPDs?


       - Andy.
  • 0

    broadgage:

    My view is that failures of this nature are acceptable provided that they remain very rare. It is not worth spending vast amounts of money to mitigate something that happens a dozen times a century.

    If however this sort of failure becomes more common, then something needs to be done. This "something" could consist of large flywheels, batteries, or the prompt cutting off of customers who have agreed to this in return for a preferential tariff.




     

    Agreed. If failures such as this remain rare then it would make far more sense to spend money on the means to cut off lower priority customers first and ensure that essential services such as the transport networks and hospitals etc remain only a last resort.
  • 0

    Jam:

    National Grid have released their interim report. I've attached it but you can also download from National Grid's website The magnitudes are exceedingly similar to the event in 2008 I mentioned before. The frequency plot does suggest rapid swings in frequency at the time of the event but while some local generation was lost due to the HV earth fault, most of the connected embedded generation stayed connected, suggesting that the ROCOF was within acceptable limits; most of the generation that was lost was as a result of issues within the plants themselves (some which have yet to be explained). The load shedding occurred 85 seconds after the event, while frequency response services (including batteries contracted for this very purpose) went from 0 to 650MW in less than 10 seconds... The story is not so much about inertia as it is an imbalance in supply and demand due to the sudden loss of supply.




    I'm not sure I agree with your take on the report.  The report states that 500MW of embedded generation at the distribution level was tripped by Loss of Mains protection.  That 500MW is a substantial part of the generation shortfall.  The most common Loss of Mains protection is RoCoF protection and most of it would have been set to trip at 0.125Hz/s under the G59 regime.  The revised G99 regime increases the trip setting for plant commissioned after May this year.  Maybe that change in setting should also be required retrospectively for plants commissioned before this year.


    After all, the G59 document (2015 revision) states, "With the changes in generation mix expected over the next decade it is unlikely to be economic to contain all frequency excursions within 0.125Hz/s.  Therefore the maximum system RoCoF which may be experienced for the maximum loss of generation infeed or block of load will rise over time.  Studies indicate that by 2023 this may be as high as 0.5Hz/s, and that even higher levels may be experienced after 2023."


    So this is a story about low inertia, low inertia exacerbating what should have been a survivable event.

  • 0

    John Russell:




    Jam:

    National Grid have released their interim report. I've attached it but you can also download from National Grid's website The magnitudes are exceedingly similar to the event in 2008 I mentioned before. The frequency plot does suggest rapid swings in frequency at the time of the event but while some local generation was lost due to the HV earth fault, most of the connected embedded generation stayed connected, suggesting that the ROCOF was within acceptable limits; most of the generation that was lost was as a result of issues within the plants themselves (some which have yet to be explained). The load shedding occurred 85 seconds after the event, while frequency response services (including batteries contracted for this very purpose) went from 0 to 650MW in less than 10 seconds... The story is not so much about inertia as it is an imbalance in supply and demand due to the sudden loss of supply.




    I'm not sure I agree with your take on the report.  The report states that 500MW of embedded generation at the distribution level was tripped by Loss of Mains protection.  That 500MW is a substantial part of the generation shortfall.  The most common Loss of Mains protection is RoCoF protection and most of it would have been set to trip at 0.125Hz/s under the G59 regime.  The revised G99 regime increases the trip setting for plant commissioned after May this year.  Maybe that change in setting should also be required retrospectively for plants commissioned before this year.



    So why didn't all generators connected in accordance with G59 trip out? None of the sites I have enquired after experienced a "G59 trip" as a result. I would suggest that National Grid is trying to say that they dropped out as a result of the earth fault, either because the associated section disconnected or because other LOM protections (UV/OV/UF/OF or possibly intertrip) operated during the resulting transients. I agree that it would be good to know though.


    Also, there are plans to change existing SSEG's settings: Distribution Code website

  • 0

    So this is a story about low inertia, low inertia exacerbating what should have been a survivable event.



    Even if all the generators had been conventional rotating electromagnetic machines - how could the grid have survived a sudden loss of nearly 1.9GW of generation? (at least without load shedding). I can see that Inertia might keep things going (at a gradually declining frequency) for a short while - but for how long before the frequency (and voltage) drop to unacceptable levels? My gut feel is that it wouldn't be very long (a few tens of seconds - or a minute or two perhaps at a complete guess) -  but surely that's nothing like long enough to bring conventional grid-level generation on-line out of the blue to compensate?


      - Andy.
  • 0
    Re Andy's post, I think the issue may be whether the second unit would have tripped if it had been a high inertia generator.

    The reasons two units tripped hasn't been explained as far as I know.
  • 0
    “So in short a lightning strike on a distribution line had three effects”


    No. The lightning strike was on a Transmission line, which has no directly connected consumers. 


    “1. Tripped out the protection for that line (as intended), losing both loads and embedded generation in the area served, then the protection autoreclosed within 20s restoring the loads, but embedded generation has to wait longer before it's allowed to reconnect - so the load on the national grid increased by the amount of the embedded generation (500MW they say).”


    No loads or generation were lost as a result of the line opening. Losses were as a result of the under-frequency excursion. 


    “2. The spikes from the strike also caused two of the three interconnectors from a wind farm to trip out (not expected) - losing 737MW”


    We know that an issue has been addressed at the wind farm, the information as to what it was is not in the public domain. We do know that the interconnector to the grid was not tripped, the issue was at one or more of the windfarm’s substations. 


    “3. The spikes from the strike also caused one steam turbine at Barford Gas Power Station to trip out (not expected) - losing 244MW”


    The reasons behind the trip of the steam turbine at Little Barford are not currently in the public domain. 


    “The loss of the steam turbine knocked the combined cycle system out of kilter, so causing the two associated gas turbines to trip out within 90s - losing a further 397MW”


    Yes, in so far as with the steam turbine tripped, the steam produced by the gas sets has nowhere to go, and either trip on high steam pressure, or need to be tripped by an operator. 


    “So the grid was then out of balance to the tune of 1.9GW - while the grid has 'reserves' ready for only 1.0GW - enough to compensate for the loss the largest single generator - so the DNO's systems started load shedding.”


    Yes. It is easier to shed 5% of the load and recover from it, which took under an hour on the day, than lose 100% and be in a true black start scenario. That is a difficult and time taking scenario, and has never (yet) been done in the UK. 


    “Is it just me or does 500MW of embedded generation on one single distribution line sound like a lot”


    The 500MW of embedded generation that tripped would be spread throughout the country. 


    Regards,


    Alan.
  • 0

    AJJewsbury:


    Even if all the generators had been conventional rotating electromagnetic machines - how could the grid have survived a sudden loss of nearly 1.9GW of generation? (at least without load shedding). I can see that Inertia might keep things going (at a gradually declining frequency) for a short while - but for how long before the frequency (and voltage) drop to unacceptable levels? My gut feel is that it wouldn't be very long (a few tens of seconds - or a minute or two perhaps at a complete guess) -  but surely that's nothing like long enough to bring conventional grid-level generation on-line out of the blue to compensate?




    No, it isn’t long enough to bring on a set from cold, but there are other strategies that could have worked, for example setting all running sets to maximum, requesting additional power from the European Interconnections, and even voltage reduction. What we had on 9th August was not sufficient to organise all the things that would have been needed. Additional inertia would have bought the grid more time, and may have bought sufficient time.  


    Regards,


    Alan. 

  • 0

    Alan Capon:



    ...there are other strategies that could have worked, for example setting all running sets to maximum, requesting additional power from the European Interconnections, and even voltage reduction. 



    As I understand it, the 'reserves' the grid has (the 1GW figure being bandied around) is the additional power that is available by running all sets to maximum plus any additional power that can be drawn from the European Interconnections instantaneously. The whole point here is that these strategies were implemented and did not work. Reducing the voltage is a very tricky proposal and not one I would suggest. We have a defined supply voltage of 230V which I believe has a tolerance of +10%/-6% and additionally installations may have their own voltage drops from the supply to consumer of another 5%. If the voltage is dropped much below this -6% plus -5% we are getting dangerously close to the the level at which contactors may start to drop out, causing even more of an issue than a straightforward loss of supply.

    Alasdair

Community Rules & Guidelines