SYNAPS Phase 1 demonstrated that our AI-powered SYNAPS system can reliably predict faults on our networks. Phase 2 will develop next generation sensors ready for connection to our low voltage network, providing a fault location within 5 metres before they have occurred and preventing power cuts.
What is the project about?
We are working with Scottish and Southern Electricity Networks and Powerline Technologies Ltd to build on Phase 1. This phase aims to:
Validate the technology further
Increase the accuracy, location and speed of estimation locations
Provide a real-time interface fully integrated into our IT system
Investigate long-term fault evolution
Improve the technology and build towards implementing it in ‘Business as Usual’
How we’re doing it
SYNAPS Phase 2 will be organised in two trial stages:
A ‘basic’ fault location trial for four to six months trialling two sets of our current prototype sensors.
An ‘advanced’ fault location trial for six to eight months trialling ten sets of our next generation sensors.
Through both stages, we’ll be gathering data about how SYNAPS performs in various fault conditions and take the technology to the next level.
What makes it innovative
Synaps Phase 2 extends our original project, which was the first time that leading-edge AI technology and ‘big data’ have been used to improve network reliability. If the trial is successful and rolled out across the network it could potentially help halve the number of power cuts.
Using machine learning by analysing waveforms from different locations to predict faults is a first on our network. It will help us be proactive like never before, replacing expensive and time-intensive methods with automated technical knowledge.
What we’re learning
Accuracy and speed of fault location
Recorded evidence that calibration of the SYNAPS sensors when they are installed contributes to better fault location.
Fault location provided by the SYNAPS solution has proven to be accurate following independent forensic examination, by RINA, of a 5m cable splice.
Improved the speed with which fault/defect location can be determined, from three weeks in SYNAPS 1 to less than 1 day in SYNAPS 2
SYNAPS on radial and meshed networks
Demonstrated the solution can detect network disturbances on radial and meshed networks.
Long-term fault evolution on faulty cables
Demonstrated that long-term monitoring of cables can identify faults as they evolve. For example monitoring a cable section over a few months.
We use cookies to give you the best possible experience when using our website. Many of these cookies are essential to the efficient operation of this site.
