Ahmed Hammada, a load balancing engineer with National Grid, began by giving an overview of the UK’s electricity transmission system. It is owned by two operators in Scotland, SP Energy Networks in the south and SSE in the north, and National Grid in England and Wales. Regardless of the ownership of the infrastructure National Grid has the responsibility of managing the whole network through the UK’s Electricity National Control Centre (ENCC). National Grid is expected to become two legally separate entities under Ofgem proposals, undertaking grid control and infrastructure management.
The UK’s transmission network was described as operating at 400/255 kV and having High Voltage DC (HVDC) interconnections to Northern Ireland, Ireland, France and the Netherlands with a connection to Belgium planned. The interconnection between the English and Scottish networks has been recently reinforced by the Western Link, an underwater HVDC link, between Ayrshire and The Wirral.
The various factors that influence the system load were described, cyclical events like the time of year, day of week or time of day, the forecast-able, such as weather or national celebrations, and the unpredictable. All of this needs a second-by-second balancing of the system. It was explained that there was a statutory duty to maintain the supply frequency at 50 Hz ± 0.5 Hz and effectively this was a measure of whether the system is in balance or not, (a heavily loaded generator slows down, a lightly loaded generator speeds up and that is reflected in the supply frequency). Typically a change of 1.5 GW in supply or demand will change the frequency by 0.5 Hz. Should the frequency ever exceed 52 Hz generators begin to trip automatically, (to prevent over-speeding), and below 48.8 Hz automatic disconnections will occur.
Several illustrations of seasonal and daily load changes were shown and it was noted that because of solar power the midday power demand in summer could well be less than at 4 a.m. and indeed the 3 p.m demand was often less than at 3 a.m. While demand forecasting was used the error margin between forecast and actual demand was usually in the range of 3 – 5%. Having discussed why demand and supplies were variable and why they needed to be balanced Ahmed went on to explain the methods that were available to him.
Sudden negative swings in the system balance could be countered by bringing in fast-reacting plant, such as the pumped storage scheme at Dinorwic, or hydro power stations. Load-shedding agreements with customers could also help. Gas turbines and customer-owned diesel generators could be brought online quickly and the Combine-Cycle Gas Turbine (CCGT) plant, that makes up 50% of the UK capacity, reacts moderately quickly. Nuclear plant was expected to supply base load and not to take up fluctuations, although in theory the option was available but at a very high offer price. For positive swings suspending supplies from wind turbines was becoming more common and similar arrangements for solar farms were being considered. These balancing operation took place against an order of merit, taking into account reaction times and the relative costs, it being explained that payments were made for supplying extra power on demand and for agreed load-shedding.
There then followed an extensive discussion around the changes that might be expected in the UK network, in which the audience joined in. The increasing amount of solar and wind plant brought new problems as the output could be highly variable. The move away from large fossil-fuelled turbo generators also removed the element of mechanical energy storage that had helped system stability in the past. It was pointed out that the South West of England, at certain times, was self-sufficient using solar power alone – might networks become more local? The use of electric vehicles was briefly covered, would their owners be prepared to see their batteries used as a network energy store?
I did wonder afterwards if the second part of the title, ‘The Bit You Don’t See’, was meant the way the speaker intended. After all most of us don’t see any part of the electrical supply system, the lights ‘just come on’. However I hadn’t realised how complicated the system had become, not just in an engineering sense, but in a commercial way, with payments being made for not using electricity as well as for supplying it. Another complication, which was covered but perhaps not explicitly spelt out, was that a lot of generating capacity is being added to the distribution network rather than to the transmission network. What had been a customer of National Grid can now become a supplier and worse than that a supplier outside the grid control. (I am thinking here of thousands of domestic solar arrays suddenly leaping into life as the cloud cover breaks). Would it be possible, (desirable?), to have domestic solar inverters cut-out outside of set frequency limits?
An interesting point was made during ‘questions’, not only might electric car users not appreciate their batteries being used for ‘balancing’ they might also regret the ‘stealing’ of limited battery charge/discharge cycles. Mention was made of the mechanical inertia that the steam turbo generators provided to the system. I am no expert here but I make a guess of about 44 MW-seconds for a 600 MW set dropping 0.5 Hz. Does anyone know better?
This was certainly a thought-provoking talk. The present, as revealed, is certainly more complicated than most of the public realise and the future looks to be even more so.
Contribute Your Own Solar Power Readings
