 |
|||
|
July 2008 |
Issue #41 |
||
     |
|||
|
|
|||
Did the Lights go out for you?
Duncan Jack, lecturer on the UK Electricity Markets Course, and lead consultant for the MJMENERGY TECCS Electricity service, looks at surprising recent events in the UK electricity market. 27 May saw one of the most dramatic events in the modern Electricity Supply Industry. So significant, that the technical workings of grid control even made it into the national broadsheets. But there the speculation ended with so little mention in the evening’s radio and TV bulletins that the conspiracy theorists went into overdrive. The system has been unusually short over recent months counteracting the longstanding industry propensity to contract long by 1-2%. Commodity prices are high, perhaps this is driving the industry to run closer to the wind, and April saw the system at its shortest since NETA began in 2001. But crucially on the 27th May there were no system warnings in place, not even our old favourite the NISM (Notice of Insufficient Margin), and the day began uneventfully. Margin, the main measure by which National Grid quantifies its ability to respond to system events was healthy and apparently well within their obligations as set out in the Security and Quality of Supply Standard (SQSS) and Grid Code. One of the plants singled out for attention was Longannet Unit 1 which was the first unit being brought back at Longannet after the whole station had been on outage for its summer maintenance. By all accounts it was struggling to come to full load as is often the case after units have been on prolonged outage. But it was by no means the only plant experiencing difficulty that morning. Cottam had already been pulled earlier in the morning and there was a succession of small to middling plant losses all morning. But in that respect this is no different to any other day. But finally at 11:34, Longannet 1 “could nae take it anymore[1]” and was taken off the bars. Contracted primary response kicked in followed by the requisite secondary response and all was well with the world. Until 11:36 that is when, in an unrelated event (in the press British Energy were initially quoted as suspecting a fault with a sensor on the steam circuit) the full load of Sizewell B’s two generating units was tripped off. And if there is one thing we already know about nuclear generation, the physics, engineering and economics all combine to mean that it comes in large lumps. In Sizewell’s case, 1175 MW of a lump. Since it went operational Sizewell B has represented the largest potential lost load on the system and the Grid plan to meet this particular event. Indeed as attendees at our Introduction to UK Electricity Markets courses know, May 2003 gave them a dry run for controlling the system under such circumstances. Again the frequency response responded manfully to the job. But then the system was hit by a third, as yet unexplained frequency drop. At this point, primary and secondary response had been all but depleted by what Americans might call a “double dip” into the reserves and the frequency dropped dramatically to 48.795 Hz. The third decimal place is particularly important for consumers as 48.8 Hz is the trigger for the first tranche of automated Low Frequency Demand Disconnection (LFDD). 27 May was the first time in 20 years that LFDD had operated but it meant that 581 MW of demand was automatically disconnected from the Grid at points as diverse as Cheshire, East Anglia, Lincolnshire, London, Merseyside and Teesside. In a more controlled measure Grid Control also initiated Disconnection by Instruction which, rather than complete disconnection reduces the voltage supplied to customers – it is this practice that leads to “brown outs” as the lights dim but don’t completely go out. A further 1200 MW was achieved through this measure. A difficult lunchtime was followed by a gruelling afternoon for Grid Control as significant plant losses continued throughout the run up to the tea time peak. An HRDR (High Risk of Demand Reduction) system warning was issued at the time of the original triple hit and extended during the afternoon, followed in Period 32 (14:30-15:00) by a DCI (Demand Control Imminent) system warning. There’s only one more system warning in the bag, the rather innocuous sounding Risk of System Disturbance. The system warnings were not lifted until teatime. And it was not until mid evening that Sizewell returned to the scene and began the slow careful crawl back to load that nuclear sets must undertake. It wasn’t back to full load until 29 May – two days after the incident. Much searching of the record books suggests that we have to go back to 1965/66 - when national frequency dropped to 48.72 Hz – to find a national event as severe as that experienced in May. A more recent contender for the title of lowest frequency was also accompanied by a system split that meant that the effect was not felt nationally. What caused the mysterious third dip? Certainly it is unlikely to be a demand feature and National Grid has full access to all the generation metering so they could easily identify if a third genset fell over during the same event. This suggests a more complex event perhaps coupled with problems on the grid itself that need to be fully disentangled. And is this a vision of the future? Will this sort of event become more common in a system with an increased reliance on both “chunky” nuclear and ephemeral wind?
This account builds on our first view of
events presented at our UK Electricity Markets public course and is
supplemented by details presented by National Grid at its recent
Operational Forum. [1] Apologies to trekkies everywhere. |
|||
|
MJMENERGY LTD have provided all the information in this newsletter
free of charge to anyone who wishes to read it. We cannot be held
responsible for any inaccuracies although all information is
believed to be correct at time of publication. Whilst articles
published in this newsletter often carry a particular point of view,
publication of them does not imply that we necessarily agree with
them. Anyone wishing to contact the editorial team with regards to
any of the above articles should email: editor@mjmenergy.com, or
phone +44 (0) 845 299 7072. |
|||
