There are broadly three types of maintenance strategies, explains Javier Ortego, director of BlueBox Technology at Ampacimon:
Reactive or breakdown is a strategy of responding to a failure after it has occurred and when repairs are needed to restore function.
A preventative or scheduled maintenance strategy requires periodic inspections and regularly timed interventions.
Any defects are corrected as they are detected or components are replaced at scheduled intervals.
A preventative strategy is designed to correct failures before they occur but can result in potential faults being missed or elements being replaced when they are still functional.
Predictive maintenance is a data-driven, proactive strategy that relies on continuous data recovery and analysis to accurately forecast potential failures before they occur and allow timely intervention.
While partial discharge monitoring can be used as part of a preventative maintenance strategy with more sophisticated analysis partial discharge monitoring can also be deployed as in important element in a predictive maintenance operation.
Ampacimon, for example, has developed its GridLife suite of grid monitoring tools including PDEye.
This serves as a central monitoring platform for all the different sensors and monitoring units installed across the grid by detecting defects in insulation using permanently installed equipment such as sensors.
Installed on the cloud or on-premises it can be connected to the asset management system to accurately monitor the system for partial discharge and enable automated diagnosis across all the various asset classes, including cables, transformers, substations and switchgear, generators and motors, and gas-insulated substations.
Developed for optimal reliability, PDEye automatically generates instant real-time warnings from this sensor data with a 98% accuracy.
This level of precision reduces maintenance costs but coupled with an advanced artificial intelligence analytics tool, the platform not only identifies faults and their location but also provides an evaluation of any detected defects.
This AI modelling provides a diagnosis for the technical teams, identifying the defect type, any patterns, their criticality, and other parameters allowing it to categorise multiple defects automatically through clustering. This analysis considers any localisation, sensor ratio, wave parameters, and the phase-resolved partial discharge pattern and delivers an accurate list of defects from just a single measurement.
The system recognises these diverse defects in all insulation types, including XLPE, air, oil, or SF6. This precise AI modelling not only reduces the need for expert analysis but allows non-expert technicians to rapidly assess the condition of any assets and quickly plan and execute any necessary preventive actions. It can therefore enable a predictive maintenance approach to be adopted.
The future of an ageing grid
By detecting and taking action to address potential problems before they occur, grid reliability is improved. In addition, by acting early the cost impact of any emerging problems is mitigated.
For an ageing asset base where reliability is already likely to be affected, advanced grid-enhancing technologies like partial discharge monitoring coupled with AI analytics are tools that serve a multitude of important functions.
Indeed, a recent Credence Research report on the Middle East Grid Modernisation Market found that the market is anticipated to reach US$2.6bn by 2032, at a CAGR of nearly 15%.
It’s a market largely driven by increasing investments in smart grid technologies and digital transformation in the power sector as well as the influences of renewable energy integration.
Partial discharge monitoring allows companies to be aware of the health of their assets and make better decisions about maintenance and repair.
And, while it is worth noting that any defect that does not originate in the main insulation is not detectable by partial discharge monitoring, it can nonetheless serve as an important tool in the grid operator’s arsenal.
This is the last part of Ortego's op-ed. Click here to read the first and second parts.