A breakthrough platform co-developed by Ampacimon, Elewit, and Red Eléctrica has enabled the identification of functional anomalies in early-stage critical assets in power transmission networks, such as the insulation on electrical assets such as power cables, which allows for proactive correction before failures occur.
This cutting-edge solution and initiative have been developed with the field of predictive maintenance front of mind, arising from the possibility of incidents a network may suffer, affecting power supply quality and continuity.
The new innovative, modern technology, PDEye, has been designed in response to potential incidents suffered by electrical cable power networks, which can, in turn, affect entire power assets, such as transformers, transmission lines, and substations, all vital to the stability of grid efficiency. Therefore, it is crucial to detect these potential failures early with preventative solutions to ensure they don’t escalate further, with operations and maintenance personnel able to anticipate problems and act leading to a reduction of unplanned grid outages.
Ampacimon, Elewit, and Red Eléctrica have combined their years of industry experience to jointly launch PDEye in 2021 as an innovation project in-house at first, then after hugely positive results, scaled to launch as a product to the market as part of the inventory of tools at Redeia maintenance unit – an industry first!
PDEye platform operation: monitor, detect, diagnose, respond
PDEye can monitor critical variables and detect small discharges related to manufacturing or installation defects in the insulation of cables before severe incidents (partial discharges), which can cause bad cable damage and subsequent faults – tripping events interrupting power and requiring emergency repair work.
Partial Discharge diagnosis, whilst highly useful, has often been considered ‘black magic’ – this is because the specialist engineer must be involved in the diagnosis to understand the complex signals from the sensors-similar to how a doctor can look at an X-ray that confuses most people and diagnose an illness. In this case, ‘Machine Learning’ is used to train a computer to diagnose the same illness from X-rays when it has been trained using thousands of previously diagnosed X-rays. PDEye uses ‘Machine Learning’ to build on the years of knowledge of the PD expert and automatically cluster and diagnose cable faults with an exceptionally high success rate of over 90%.
Therefore, PDEye can diagnose and detect early any manufacturing or installation defects of a range of components in electrical network assets, as well as control the gradual deterioration of components throughout their lifecycle, external anomalies, damage, and exposure to heat cycles, making it useful for monitoring the operation of cables, transformers, terminations, electrical switchgear, and other components in various network assets.
Improved efficiency and lifecycle of critical assets
Planning based on the real-time state of the equipment, and clustering using ‘Machine Learning’ AI advanced technology, offers network operators a real advantage in improving overall asset efficiency and useful life, which minimises downtime and the associated losses.
Furthermore, it enables power cable and power grid resilience and adaptability from its ability to use data to monitor, detect, diagnose, identify, and address/respond to anomalies early, providing valuable information for strategic decision-making, such as infrastructure investment planning and electrical grid optimisation.
Image source: Courtesy of Ampacimon; stock image for illustration purposes only
Read more about Ampacimon here