HV Pilot (High Voltage Phase Identification)

Project partners: CGI and Loughborough University

Project HV Pilot will be using smart meter data and helicopter surveys to determine a method to identify how the low voltage (LV) network is connected to the high voltage (HV) network and how balanced or unbalanced the network is. This is of particular interest in rural areas where HV overhead lines have lots of single-phase sections of HV network, increasing the chances of the network being unbalanced. If successful, the project could enable cost-savings to networks, reduced voltage issues for customers and an increase in the number of connection opportunities.

Problem(s) 

Electrical power distribution on overhead 11kV networks often includes instances where a three-phase network supplies single-phase sections of line and/or single-phase transformers. The exact connectivity, identifying how the phases are connected is often missing from   network models. This transition greatly impacts network efficiency and reliability. Long-distance electricity transmission favours three-phase power for its efficiency and higher capacity over single-phase power. However, most residential and light commercial consumers use single-phase power. Higher loads connect to one phase of a three-phase cable, balancing customers across phases. Rural areas use single-phase feeder mains, needing only single-phase transformers. HV overhead spurs for single-phase transformers typically use a two-wire network on the same conductors of three-phase HV feeders. 

Accurate modelling is crucial for predicting network changes, demand increases, and renewable integration, impacting performance. Inaccurate models risk inefficiencies and failures. Single-phase load variability challenges load balancing, causing voltage fluctuations and increased equipment wear. Managing unbalance in rural LV networks faces challenges due to insufficient three-phase to single-phase documentation, worsened by low-carbon technologies like heat pumps and EV chargers, increasing losses and voltage variations. Optimally balanced loads could reduce losses and enhance capacity for these technologies.

Addressing data gaps requires thorough data collection on single-phase connections. Site visits to establish phase connectivity are costly, especially for pole-mounted transformers. The SMITN (Smart Meter Innovations and Test Network) project shows smart meter data can identify phase connectivity for single-phase customers. HV networks benefit from advanced grid management systems capturing real-time data for dynamic modelling. The HV Pilot project aims to determine HV phase connections of single-phase transformers, using technology and advanced data management, including digital sensors, monitoring systems, and images identifying transformer phase connections.

Method(s) 

The HV Pilot project aims to find solutions to accurately determine the phasing of high-voltage feeders and associated single-phase transformers in rural areas. This will be achieved by managing data from various sources like digital sensors, monitoring systems, and images. The goal is to enhance transformer phase connection reliability and precision for accurate network modelling and to determine the distribution network's hosting capacity for new LCT connections. 

The earlier SMITN project developed methods for determining LV phase connectivity, but gaps exist at the HV network level. Recent advancements in network visibility, such as monitoring equipment at distribution transformers and smart meters, facilitate real-time data collection. Imagery from satellites and aerial photography enhances infrastructure understanding, offering ways to address network unbalance and improve LV network efficiency and reliability. 

The project has five work packages: data investigation, infrastructure development, phase identification using smart meter data, phase identification using imaging data and machine learning, and dissemination and trial specification. Each package includes deliverables like data assessment, infrastructure design, algorithm selection, and report creation. The final deliverable is a SIF Beta trial application with a cost-benefit analysis. 

• Work Package 1 involves organising workshops to evaluate network data, including phase indicators on schematic diagrams, LiDAR data, and helicopter images, to create a network model for area selection 
• Work Package 2 identifies hardware, software, and interfaces for data collection, integrating it for analysis 
• Work Package 3 tests algorithms for phase identification using smart meter data, evaluates data issues, and documents learning outcomes 
• Work Package 4 develops machine vision systems for phase identification, assessing methodologies and performance, and addressing data issues 
• Work Package 5 compiles a Combined Findings Report, identifying optimal phase identification approaches, establishes a business case, and submits the SIF Beta application

Scope 

The scope of the HV Pilot project involves investigating and applying algorithms that use various data sources for the following use cases. 

1. Accurate determination of high-voltage (HV) feeder phasing and associated single-phase transformer phasing in rural areas 
2. Integration and management of data from digital sensors, monitoring systems, images, and existing databases to enhance the reliability and precision of transformer phase connection information 
3. Utilisation of smart meter data and machine vision for phase identification, ensuring the accuracy and effectiveness of these methods 
4. Development of infrastructure to support data collection and processing, enabling efficient analysis and interpretation 

If the project is successful, it will lead to improvements in data quality and availability, which will then allow for:  

1. more accurate network planning by modelling unbalanced three-phase networks   reduced energy losses through better phase selection for new connections  
2. enhanced load balancing, leading to fewer voltage complaints and reduced fuse operations from unbalanced phases 
3. improved capacity for low-carbon technologies (LCTs) such as heat pumps, electric vehicle chargers, and photovoltaic systems  
4. better targeting of network monitoring and reduced costs associated with manual surveys  
5. identification and correction of data errors ahead of need  
6. improved use of network data by third parties, contributing to a more efficient and reliable power distribution network 

The financial benefits from these improved outcomes are estimated to be approximately £2.5 to £2.8 million per annum within NGED’s licence areas, made up of: 

• savings from reduced voltage complaints and improved network balancing 
• reduced network losses and customer bills 
• savings from reduced manual survey costs for phase confirmation 
• enhanced network planning and reduced costs associated with inaccurate fault location identification

• Objective(s)
  1. Explore solutions for accurately determining the phasing of high-voltage feeders and associated single-phase transformers in rural areas.
  2. Investigate viable methods for accurately determining the phasing of high-voltage feeders and single-phase transformers in rural areas  Integrate and manage data from various sources, including digital sensors, monitoring systems, images, and existing databases, to enhance the reliability and precision of transformer phase connection information 
  3. Develop algorithms using smart meter data to identify the connection between low-voltage (LV) and high-voltage (HV) systems 
  4.  Identify a methodology to address high-voltage connectivity gaps, particularly for single-phase transformers, to maintain an efficient and balanced network 
  5. Leverage recent technological advancements in network visibility, such as monitoring equipment at distribution transformers, smart meter data, and imagery from satellites and aerial photography, to address network unbalance issues 
  6. Apply the algorithms developed for the test network to real-world scenarios 
  7. Assess the performance of the algorithms and identify factors that affect their accuracy
  8.   Capture the learning from the project and disseminate it to interested parties
• Success Criteria
  • Successfully assessed and evaluated network data to identify the project sample areas 
  • Analysed helicopter images to determine their usability and relevance for applying machine learning techniques 
  • Compiled comprehensive reports on network data assessments, visual data analysis, and the outcomes of area selection 
  • Identified and integrated suitable hardware, software, and interfaces to facilitate efficient data collection, processing, and analysis
  •  Thoroughly tested and evaluated algorithms, to ensure they met the specified use case requirements
  •   Produced detailed reports on selected algorithms, the development and optimisation of methods, and the learning outcomes from applying these algorithms 
  • Developed a robust machine vision system for phase identification, complete with thorough methodologies, techniques, and performance analyses
  •  Clearly summarised project findings and construction of a compelling business case and successfully developed a beta application
• Potential for New Learning

  The knowledge dissemination activity will cover the learning from the practical implementation of the algorithms to identify the HV phase connections of single-phase transformers, particularly in rural areas. This will outline how well the algorithms worked and what factors affected their success rate, such as the quality and availability of smart meter data and other data sources. 

  The learning from the integration and management of data from various sources, including digital sensors, monitoring systems, and imagery, will identify whether the additional complexity of combining these data sources results in improved accuracy compared to simpler methods. It will also detail the accuracy of the phase connection information and how it was validated using different data sources. 

  The learning will also include insights from the development of machine vision systems for phase identification, how these systems were validated to ensure they were suitable for use, and any practical issues encountered during the implementation. 

  Learning will be disseminated during the project externally via: 

  • One or two webinars 
  • LCNI (Low Carbon Networks & Innovation) Conference 
  • Innovation Showcase 

  These dissemination activities will ensure that the knowledge gained from the project is shared with other Distribution Network Operators (DNOs) and relevant stakeholders, promoting the adoption of best practices and further innovation in the industry.

• Monthly updates

  • {{update.title}}

  Show more