SCOPE
Use of hydrogen as an energy source
Maturity Level
TRL 5-6
Development Level
Proof of concept
Protection Level
Knowledge register
Predictive maintenance and failure detection
Entity:
Unlike corrective maintenance, predictive maintenance makes it possible to programme the actions to be taken before the failure becomes serious and the system’s operation is interrupted in an unplanned manner. The application of machine learning techniques makes it possible to detect deviations in the behaviour of equipment with respect to normal operating patterns. Using domain and historical knowledge about failures, observed anomalies are related to potential failure symptoms and alarms are generated by maintenance personnel so that they can take corrective actions to avoid system inefficiencies or interruption of operation due to a total shutdown.
Challenges met
- Avoids unscheduled service or equipment interruptions
- Detects malfunctions and minimises inefficiencies
- Reduces the replacement of components that have not yet reached the end of their useful life
Scope of application
Maintenance of facilities, including hydrogen production equipment
Main publications
Related projects
- SPHERE – BIM Digital Twin Platform
- SOLARADAPT – Active and intelligent management of solar photovoltaic energy production and local consumption in flexible power grids
Other related technologies
