Distribution Network Operators (DNOs) experience faults on their electricity distribution networks, which result in Customer Interruptions (CIs) and Customer Minutes Lost (CMLs). Most CIs and CMLs are incurred on the high voltage (HV) network. DNOs implement a number of measures to reduce the amount of CIs and CMLs incurred, for example through switching via automation and or remote control, use of protection relays to identify faults and minimise impact. However, these measures mostly only address scenarios where the fault has already materialised. Further improvements in network performance and reduction in operating costs could be achieved if DNOs are able to monitor key network characteristics, e.g. voltage and current in real-time, and carry out interventions (e.g. asset or component repairs) before a fault materialises. Monitoring network characteristics in real-time presents some practical challenges and considerations. For example:
• What are the typical network characteristics that are identifiable before different types of faults?
• How would the location of the emerging fault be identified?
• What are the operational processes and steps that would need to be followed to successfully pre-empt an emerging fault?
This project aims to test a solution, “Distribution Fault Anticipation” (DFA), to monitor feeders to pre-empt faults. The DFA solution consists of a disturbance recorder (which can be installed on HV or 33kV feeders to monitor network characteristics) and a “Master station” (a cloud-based service which provides the secure conduit and main data repository between the DFA and the DNO). This will be trialed alongside a network analysis tool (ASPEN Distriview) and Fault Passage Indicators (FPIs) to monitor a selection of HV and 33kV feeders and expectantly identify the location of network issues before they manifest into faults.
Objectives
The objectives of the project are to:
• Validate the process of sampling network characteristics (such as voltage and current), using DFA devices, in real-time to identify pre-fault disturbances;
• Validate the process of analysing system performance, and sending notifications of pre-fault disturbances, in real-time;
• Prove the analytical capability of the DFA devices to identify different types of pre-fault conditions in real-world environments;
• Develop and validate a process to use outputs of multiple tools (DFA devices, FPIs, protection, modelling tools etc.) to identify the source and location of pre-fault disturbances;
• Develop and validate an operational process for responding to DFA outputs and carrying out repairs to pre-empt faults.
Learnings
Outcomes
The trial of the DFA-Plus solution showed that it is fully achievable to employ the technology to a range of pre-fault and post-fault use cases. In the pre-fault use case the technology was used to identify a wide range of leading indicators of network events which were indicative of emerging or underlying issues on the electricity distribution network. In the post-fault use case, the trial also showed that the solution could be used to identify fault locations more quickly.
During the project an end-to-end process was trialled which involved: reviewing notifications provided by the DFA-Plus devices, network modelling using DFA-Plus-sourced outputs, field investigations to confirm issues highlighted by the DFA-Plus. This process showed that the DFA-Plus could reliably detect and classify waveforms of network events and send pertinent notifications to alert users. These alerts were a learning from the project also showed that these outputs when modelled and used alongside other information (e.g. network data from FPIs, auto reclosers and weather), could provide reasonably accurate insights.
Nortech FPIs were successfully used to identify fault locations on the feeders that the DFA-Plus solution was deployed on and facilitated quicker fault locations for the post-fault use cases. However, the attempt to use the FPIs to detect pre-fault network disturbances, could not be proven due to the large mix of both normal and abnormal event activity at much lower triggering thresholds on the circuits that were monitored. Learning from the project also indicated that the varied network characteristics and types of issues observed on different feeders would have meant different threshold settings for the FPIs on each feeder.
A summary of the network events, that were identified during the project, is avilable on request. Details of a selection of these events are given in the final report (“HV Feeder Monitoring Final Report v1.0”). The ability to identify these types of network issues provides the opportunity to pre-empt major faults, and to locate faults more rapidly on HV and 33kV networks.
A summary of four sample incidents witnessed during the project, that highlight the opportunities to pre-empt faults or find faults more quickly, is given below:
· Incident 1 (Faulted cable termination on the Canterbury 06 feeder in UK Power Networks)
o In September 2020, there was a failure of a cable termination on the Canterbury 06 feeder in UK Power Networks. A review of the notifications from the DFA-Plus solution, showed groups of related network events more than two months before the failure of the cable termination. Analysis of each of these events, identified a failure propagation path from phase-to-earth, through phase-to-phase-to-earth, through phase-to-phase, with a delayed commencement of the earth fault component. This was typical of other subsequently observed cable termination failures on other feeders.
· Incident 2 (Fault Induced Conductor Slap on the on the Canterbury 06 feeder in UK Power Networks)
o The DFA-Plus devices installed on the Canterbury 06 feeder in UK Power Networks, detected a series of seven events on 17 May 2021. The shape and pattern of the waveforms indicated that the cause of the events was vegetation-related conductor clashing. Field investigations identified a section of overhead line which matched one of the two probable locations modelled by the ASPEN FLT tool, after the DFA-Plus had detected the events.
· Incident 3 (Line down event on the Calne E8L5 feeder in Scottish and Southern Electricity Networks)
o There were two DFA-Plus devices monitoring the Calne E8L5 feeder in Scottish and Southern Electricity Networks. One of these DFA-Plus devices was connected in line with the protection wiring and the other via interposing CTs. On the 28 of October 2021, both DFA-Plus devices logged six identical events, the first four events within two minutes of each other. The arcing characteristics detected and the transition from phase-to-phase to phase-to-earth indicated a line down had occurred. A field investigation revealed a line down event had occurred with c.570m of the modelled location from ASPEN FLT tool.
· Incident 4 (Suspected third party damage to an overhead line)
o On the 13 of October 2022, the DFA-Plus devices on the Wootton Bassett E4L5 feeder in Scottish and Southern Electricity Networks, recorded a single short duration reclose event on the feeder. The location and cause of this type of event would usually be unknown. The characteristics of the event indicated an immediate “A” phase to earth fault, with no pre-cursor or pre-fault activity. A site investigation matched the probable location produced from the ASPEN FLT tool. The site investigation showed that tree cutting had recently been carried out and suggests that some contact would have been with the overhead line in the process.
During the trial the DFA-Plus solution was also used to carry out a post-storm review. It grouped system operations by specific fault events and causes, to assess the impact of the storm event and identify opportunities for remedial work that may be required to improve future storm resilience.
One of the main conclusions was that the successful deployment of the HV feeder monitoring solution would require a combination of:
· the installation of the DFA-Plus monitoring device and FPIs on nominated feeders;
· nomination of trained staff to review notifications from the devices and to carry out follow-up activities (including network modelling, interpreting waveform patterns, comparing data from complimentary systems, carry out field investigations);
· use of a bespoke software package called ASPEN FLT to carry out network modelling, using DFA-Plus-sourced files (PQDiff and comtrade files), to identify approximate location of network issues;
· external consulting support provided by the supplier (LORD) and the DFA-Plus manufacturer (PSLLC). This will include training, remote commissioning support and remote operational support to interpret outputs from the DFA-Plus devices and to determine potential fault locations.
Lessons Learnt
Equipment standards
Learning from the project highlighted the importance of equipment standards, particularly if products are being trialled from outside the UK or the EU. Because the DFA-Plus solution had not been used in the UK, the project team had to ensure that the equipment provided complied with the applicable ENATS standards used by most UK DNOs. The mapping exercise that was completed by the project team showed that the DFA-Plus devices had been designed to have similar robustness and reliability as protection relays.
Equipment installations
Overhead mounted FPIs (smart Navigator 2.0s) were successfully installed from ground level using long stick procedures. This reduced the installation time and costs. It also meant minimal maintenance costs if a scenario had arisen where we would have had to recover the smart navigator 2.0 to replace a sim card.
Two installation approaches for the DFA-Plus were trialled on one feeder: One DFA-Plus installation wired conventionally (in-line with the protection circuit), and the other connected via interposing CTs. The outputs observed by both DFA-Plus devices was almost identical. The DFA-Plus installation installed in line with the protection circuit was determined as the preferrable solution. This took into consideration the large physical size of interposing CTs and that the DFA-Plus devices were designed to the same standards as protection relays.
Purchases in foreign currencies
The purchase of the DFA-Plus devices and the support service was in US dollars; the key supplier was based in New Zealand and the DFA-Plus manufacturer in the United States of America. This meant that payments made at different stages of the project were subject to currency exchange rate variations as opposed to the fixed value of the project. To mitigate the impact of currency exchange variations, the project team and its project partner, EIC, ensured that foreign currencies were bought at the start of the project. Projects involving purchases in foreign currencies will need to consider a similar move or insist that foreign vendors specify prices in local currencies.
Importing goods
Learning from the project also highlighted the need for suppliers who are importing from overseas to be familiar with the process for clearing goods in the UK. This was not an issue on the project but the imports of the DFA-Plus devices highlighted the importance of the importer having completed the relevant clearance forms.
Pilot studies
Stage 1 of the project involved a trial of the DFA-Plus solution at the Power Networks Demonstration Centre. This approach enabled the project to determine the suitability of the hardware to be used on UK systems, review the installation process and requirements, and to test a few simulated fault scenarios before committing to more involved installations on the distribution networks. This provided the assurance that the solution could be effective when deployed on the network; this was useful for stakeholder buy-in.
Network trials
The success of network trials was possible because of internal stakeholder buy-in. Key stakeholders in each of the operational areas were briefed about the project and participated in the pilot studies to observe the potential of the technology. These stakeholders helped facilitate the coordination of local resources to carry out installations of the equipment used on the project and to carry out filed investigations.
Having multiple stages on the project each stage dependent on the success of the previous stage, also provided an opportunity for project participants to reduce their financial exposure if the project were to be unsuccessful. Each of the DNO Groups had the option to withdraw from the project if the results of a completed stage were not satisfactory. It also enabled the DNO Groups to assess the amount of resources required to deliver the project, so that they could plane effectively.
