Utility Week

UTILITY WEEK | 17TH - 23RD JANUARY 2020 | 23 Operations & Assets Regulation of distributed generators to be reviewed Ofgem has recommended a review of the regulatory compliance and enforcement regime for distributed generators aer the blackout. A significant proportion was lost because of the triggering of overly sensitive loss-of-mains protection systems. According to its final report on the incident, some generators may have been in breach of indus- try rules mandating that they be recalibrated. The regulator said the review should consider extending the requirement to hold a generation licence to those smaller generators currently exempt. The initial cause of the blackout was a lightning strike on a trans- mission line in Cambridgeshire. Although it was quickly rectified, the resulting fault is thought to have tripped 150MW of distributed gen- eration with vector shi protection systems. Faulty voltage control soware also caused the Hornsea One off- shore wind farm to cut most of its output in response to the distur- bance and shortly aerwards the steam turbine at the Little Barford power station was automatically dis- connected due a discrepancy in read- ings from its speed sensors. The combined losses caused a rapid fall in frequency that tripped between 350MW and 430MW of distributed generation with rate of change of frequency (RoCoF) protections. The Electricity System Operator (ESO) attempted to contain imbal- ance by activating its frequency response reserves and was initially successful, with the frequency briefly plateauing at 49.2Hz. However, one of the gas turbines at Little Barford was then discon- nected due to a build-up of pres- sure. Around the same time, the ESO recorded an additional 100MW increase in the system imbalance, perhaps due to the further loss of distributed generation. The grid fre- quency continued to fall, tripping at least another 200MW of distributed generation as it dropped to 49Hz. When it hit 48.8Hz, the emer- gency load-shedding mechanism was activated, cutting power sup- plies to more than one million cus- tomers. It is also estimated to have disconnected another 550MW of distributed generation, bringing the estimated distributed genera- tion losses to between 1,350MW and 1,430MW. In recent years efforts have been made to prevent nuisance tripping by phasing out vector shi protec- tion systems and recalibrating those based on RoCoF to make them more forgiving. For generators with a capacity of more than 5MW, the latter was required by a series of amendments to the Distribution Code, introduced between 2014 and 2018. The code was also amended in 2010 to lower the trigger point for under-frequency protections from 49Hz to 47Hz. But given the large volume of distribution generation that was lost, Ofgem said it is "reasonable to assume" that their compliance process "has not been effective in all cases". It said this was "particu- larly concerning" because it recently approved an amendment to the code requiring generators with a capacity of less than 5MW to alter their RoCoF settings. In response, Ofgem has proposed a joint review of the regulatory com- pliance and enforcement framework for distributed generators, explain- ing: "This review should explore options for setting and enforcing technical requirements on these gen- erators, including consideration of licensing smaller generators, which would require government action." The regulator said they should begin consulting with the energy industry in spring this year. Ofgem also recommended a review of the timetable for the Accel- erated Loss of Mains Change Pro- gramme – an industry-run scheme to help generators adjust their RoCoF settings and maintain compliance with the Distribution Code. It said the Energy Network Association should submit recommendations to the Energy Emergencies Executive Committee (E3C) by April 2020. potential losses due to nuisance tripping by distributed generators, which was already known to be a risk during such incidents. The immediate cause of the power cut was a lightning strike on a transmission line in Cambridgeshire. Within a second of the disturbance, around 150MW of distributed generation with vector shi loss-of-mains protection is thought to have tripped. Along with the subsequent loss of output from the Hornsea One offshore wind farm and the Little Barford gas station, this led to a rapid reduction in frequency, tripping a further 350-430MW of distributed generation with rate of change of frequency protection. As the losses from Hornsea One and Little Barford eventually exceeded the 1GW of frequency response held by the ESO, the power cut would likely have occurred any- way. However, Ofgem noted that a poten- tial fault was simultaneously recorded on another transmission line, which could have led to the disconnection of the generators near Saltend. The regulator said the tripping of distrib- uted generators would likely have increased the total loss from this fault to as much as 1,600MW. The report said: "This loss would have exceeded the amount of back-up power the ESO was holding, causing the frequency to drop below standards, and could have resulted in a similar power outage to the one that occurred on 9 August." Ofgem said the ESO's processes for esti- mating the impact of distributed generators "do not appear sufficiently robust given the marginal levels of system inertia and poor performance of frequency response providers on the day". It said the issue is exacerbated by a lack of granular data on operational characteristics of distributed generators dur- ing network faults. "We acknowledge that the ESO has had difficulty in obtaining accurate data on dis- tributed generation," it said. "However, in our view the ESO could have been more pro- active in raising the issue of distributed gen- eration impacts on system security with the regulator and industry parties." Ofgem said it is the internal policy of the ESO to only cover for losses from distributed generation during periods of increase to the transmission system, for example, due to bad weather. The regulator said it would review both the Security and Quality of Sup- ply Standards and their application by the ESO to see whether changes are required. The ESO itself called for a review of the standards in its final technical report on the blackout. Tom Grimwood, energy correspondent, Utility Week

• Cover