Large scale UK power cuts

[dspp]

All generators connected to the transmission system or who have a contract with National Grid are obliged to have the capability to provide governor control as you describe above. National Grid decides which generators it will require and pay to provide governor service and this selection changes dynamically through the day. Governors are normally operated with a 4% droop.
Load and frequency fluctuations are getting faster, so the systems you described for managing fast fluctuations would probably have responded in 10 to 30 seconds. National Grid is now contracting with some battery operators to deliver services in less than 1 second

Many thanks for your response. Its Interesting - I'm long out of touch. In my day conventional fossil-fueled steam plant could respond almost immediately - if the stream valve was on governor control (and was not fully open). Then as the steam pressure fell, before the boiler firing picked up, there was a dip which theoretically hydro could fill. The generators I worked on were set with a droop of around 3% - but I thought in my example that 2% would provide more easily digested numbers. Is 4% now standard?
How does solar provide governor control? And what about nuclear - I thought it ran at constant output. How does governor control work on the modern two-stage gas powered plant? Again, its a long, long time since I worked with gas turbine generators - effectively a Rolls Royce Avon turbojet who's output drove a big fan. Its only a little bit of an exaggeration to say that you got power or no power!
Its also interesting to see that local governor control of the system is still fundamental. I would have thought that by now, with widespread communications links, automatic generation changing could have been executed from Grid Control.

The Avon's I ran controlled OK

Solar, and for that matter any inverter coupled source (which includes a lot of wind) senses frequency and responds accordingly. However outside of certain parameters the inverter must be programmed to disconnect. The philosophy in this respect is changing and that is a part of what we are seeing here. It is a complex situation.

regards, dspp

[TheMotorcycleBoy]

Nuclear power station are steam powered rotating generators and, in this sense, are no different from coal, gas, water, etc /
convertors (all solid state now) ...still have means to keep in sync with the grid.

So back to the original question then. If all the individual parts can cope with lower frequency, Why does it matter? Why is it not better to have the whole system creaking along at 48Hz rather than shutting huge parts of the system down?

This is a genuine question. I'm not trying to get at you or criticise your answers. I genuinely don't understand why it can't just run slower when overloaded.

Gryff

It has to lock in phase with the rest of the generators.

go to this link
http://electricalacademia.com/basic-ele ... ectricity/

scroll down, and you'll see the 2 sinusoids where they out of phase (red and blue lines). In terms of a grid lets say only 2 generators, that would mean, wherever on the graph the sinusoids aren't touching that means there is a reasonably sized instaneous potential between the 2 sources. Hence current would want to move there against negligible load. Short circuit, heat, copper windings melting, stuff catching fire, etc.

(TBF it's much more complex than this, since customers on the grid will impose loads too, and will effect the overall impedance hence phase etc.)

Matt

(EDIT: And of course if the frequency changes then phase shifts occurs. I imagine with a bit of maths you can figure out how many cycles occur before the wave trains are almost 180* out-of-phase)

[scotia]

So back to the original question then. If all the individual parts can cope with lower frequency, Why does it matter? Why is it not better to have the whole system creaking along at 48Hz rather than shutting huge parts of the system down?

Ok - a reducing system wide frequency is caused by a mismatch between generation (too little) and load (too much)- and if this mismatch continues, the frequency will continue in a downward spiral. The governors on the generators will all have opened fully, so you unless you can reduce the load, or increase the generation by bringing on-stream new generators, you have a major problem. Hence the need to load shed until the new generation is up and running.
Also there are a number of devices that rely on 50Hz (e.g. synchronous motors) - so significant deviations from 50Hz will have other deleterious effects.

[sunnyjoe]

https://www.nationalgrideso.com/power-cut-9-august-2019

[scotia]

https://www.nationalgrideso.com/power-cut-9-august-2019

So It looks like we may get some information by the end of the week.

[supremetwo]

https://www.nationalgrideso.com/power-cut-9-august-2019

More information:-
https://www.flexitricity.com/blog/frida ... questions/

[scotia]

https://www.nationalgrideso.com/power-cut-9-august-2019

More information:-
https://www.flexitricity.com/blog/frida ... questions/

Thanks for the link. It doesn't add much technical detail to what we already know. I'm sure the size and time of the Hornsea loss will have been accurately monitored. If not it suggests their SCADA systems are grossly defective. It would have been interesting to know the size of the contribution that flexitricity's battery helped in the recovery - and how long it took to switch between charging and discharging.

[sunnyjoe]

The UK Government has announced the scope of the Energy Emergencies Executive Committee (E3C) investigation
https://www.gov.uk/government/news/busi ... power-cuts

"In particular E3C will:
* assess direct and secondary impacts of the event across GB electricity networks
* Identify areas of good practice and where improvements are required for system resilience
* consider load shedding in regard to essential service customers and prioritisation
* consider timeliness and content of public communications during the incident
* make recommendations for essential service resilience to power disruptions"

[scotia]

The E3C investigation:-
E3C will submit a final report to the Secretary of State within 12 weeks, with an interim report within 5 weeks, this will include: - an assessment of the event; - recommendations and lessons for the future; - an implementation plan to deliver recommendations; and - a published report and plan. BEIS will provide secretariat for the review.
I guess we will just have to wait to satisfy our curiosity as to what happened and when

[sunnyjoe]

The E3C investigation:-
E3C will submit a final report to the Secretary of State within 12 weeks, with an interim report within 5 weeks, this will include: - an assessment of the event; - recommendations and lessons for the future; - an implementation plan to deliver recommendations; and - a published report and plan. BEIS will provide secretariat for the review.

I guess we will just have to wait to satisfy our curiosity as to what happened and when

National Grid will submit an initial report to Ofgem tomorrow. I expect there will be pressure for both of them to make some information public.

This decision to approve changes to make small generators more resilient to disturbances has been on Ofgem's desk since 25/2/19 and it has been approved today
http://www.dcode.org.uk/assets/files/RT ... 08%20D.PDF

Probably not a coincidence.

[dspp]

A blog comment that may be of interest though it doesn't add anything that is not already in this thread

http://www.ukerc.ac.uk/news/what-happen ... -2019.html

I haven't seen any hard facts yet from the initial NG ESO report to Ofgem.

- dspp

[Itsallaguess]

A blog comment that may be of interest though it doesn't add anything that is not already in this thread

http://www.ukerc.ac.uk/news/what-happen ... -2019.html

An interesting read, and nice to see everything collated in one place in such a way.

Although the level of public and corporate disruption suffered on August 9th shouldn't be disregarded, it sounds like the lessons learnt from this incident could be wide-ranging, both from a supplier/infrastructure point of view but also from that of end-users such as the rail companies who look to have struggled after the restoration of power in many cases..

Often there's nothing like a real-life scenario to help refresh those operating procedures, after all....

Cheers,

Itsallaguess

[dspp]

Some indication that Hornsea One, which is still in build, did not respond quite as anticipated,

https://www.offshorewind.biz/2019/08/16 ... roduction/

""The relevant part of the system has been reconfigured and we are fully confident should this extremely rare situation arise again, Hornsea One would respond as required.”"

- dspp

[scotia]

A blog comment that may be of interest though it doesn't add anything that is not already in this thread

http://www.ukerc.ac.uk/news/what-happen ... -2019.html

I haven't seen any hard facts yet from the initial NG ESO report to Ofgem.

- dspp

An interesting read - but the graph would have been more interesting if the time tags of the two generation losses had been displayed.
So is the first big drop the result of one loss, and the second the result of the second loss?
Or is the first big drop the result of both losses, and the second drop is the steam generators losing steam pressure - and awaiting the firing lag?

[scotia]

Some indication that Hornsea One, which is still in build, did not respond quite as anticipated,

https://www.offshorewind.biz/2019/08/16 ... roduction/

""The relevant part of the system has been reconfigured and we are fully confident should this extremely rare situation arise again, Hornsea One would respond as required.”"

- dspp

“During a rare and unusual set of circumstances affecting the grid, Hornsea One experienced a technical fault which meant the power station rapidly de-loaded – that is it stopped producing electricity," a project spokesperson said.
Interesting choice of words. Whilst conceding that it was a Hornsea fault, are they suggesting that it was caused by the frequency droop from the Barford loss? If so its probably a case of failing to test the affect of the full range of frequency deviations.

[dspp]

At the moment the facts in the public domain do not allow us to answer the questions some of you are asking.

We don't know whether the wind farm went first, or the gas turbine.

We don't know what caused either of them.

We don't know if lightning played a part in it.

There is a lot we don't know yet, but I am sure we will find out in due course.

regards, dspp

[sunnyjoe]

the FT claims to have "findings, which were relayed to the Financial Times by people briefed on the report"
https://www.ft.com/content/8b738eac-c02 ... e555e96722

[dspp]

https://www.theguardian.com/business/20 ... kout-ofgem

excerpts,

"National Grid has blamed a lightning strike for Britain’s biggest blackout in more than a decade after it caused two power generators to trip offline.

The lightning strike was one of many to hit the grid on the same day as the 9 August blackouts but in an “extremely rare and unexpected event” it managed to bring down two electricity generators more than 100 miles apart.

The report revealed that National Grid was unable to cover the twin outages at a gas-fired power plant in Bedfordshire and an offshore wind farm off the east coast of England because it did not have enough backup.

The outages took place within seconds of each other immediately after the strike, ...

"National Grid had only 1,000MW in reserve as backup supply when the outages took place. The combined power capacity lost in the twin outages was more than 1,300MW.

In addition, the strike may have also triggered outages at a string of “embedded” generators – such as small-scale renewables and diesel farms – totalling an estimated 500MW."

.... National Grid had suffered three blackout “near misses” in as many months after similar-sized outages in May, June and July, which were all greater than 1,000MW."

Interesting. More detail eagerly awaited.

regards, dspp

[scotia]

https://www.theguardian.com/business/2019/aug/20/national-grid-blames-lightning-strike-for-blackout-ofgem

excerpts,

"National Grid has blamed a lightning strike for Britain’s biggest blackout in more than a decade after it caused two power generators to trip offline.

The lightning strike was one of many to hit the grid on the same day as the 9 August blackouts but in an “extremely rare and unexpected event” it managed to bring down two electricity generators more than 100 miles apart.

The report revealed that National Grid was unable to cover the twin outages at a gas-fired power plant in Bedfordshire and an offshore wind farm off the east coast of England because it did not have enough backup.

The outages took place within seconds of each other immediately after the strike, ...

"National Grid had only 1,000MW in reserve as backup supply when the outages took place. The combined power capacity lost in the twin outages was more than 1,300MW.

In addition, the strike may have also triggered outages at a string of “embedded” generators – such as small-scale renewables and diesel farms – totalling an estimated 500MW."

.... National Grid had suffered three blackout “near misses” in as many months after similar-sized outages in May, June and July, which were all greater than 1,000MW."

Interesting. More detail eagerly awaited.

regards, dspp

Thanks for the link - I'm still cautious about a newspaper reporting on engineering matters.
Its difficult to understand why a lightning strike should cause major generating plants a considerable distance apart to fail. Hornsea seems to be confessing that they know why they failed - and they have corrected the problem. I don't know about the Gas Generator - but again there must be a defect which they will need to resolve. But the apparent size of the embedded generation that was affected is surprising - and must be a nightmare to manage.

[taylor20]

For info, NG's initial report:

https://www.nationalgrideso.com/documen ... 1/download