Explanation artificial intelligence-based condition monitoring for tidal energy turbines

Our reliance on fossil fuels primarily contributes to global warming and threatens our survival. Renewable energy is currently considered the leading solution to reduce greenhouse gas emissions. Energy extraction from the ocean's tides can help fulfill the global renewable energy demand and combat world climate crises. With 45% of Europe's citizens living in coastal regions, Ocean Energy Europe predicts that ocean energy can meet 10% of EU electricity demands by 2050. Installations at this scale will have the associated benefit of reducing the levelised cost of tidal energy (LCOE) towards a target of €100/MWh, which will make ocean energy a viable compatriot to offshore wind. To achieve this, increased performance and reliability of tidal energy devices is required. In particular, there is a need for further technology investigation and demonstration for improved reliability and efficiency of tidal turbine rotor and blades, including control and condition monitoring systems. Failure in a blade can create long downtimes. The research project will develop and demonstrate an innovative condition monitoring system for rotors and blades of tidal energy devices that provides invaluable learnings regarding performance, reliability, availability, maintainability and survivability. Machine learning and data analytics methods will be employed utilising measured data for an existing operating tidal energy device, which will improve the seaworthiness of rotor and blades, reducing the likelihood of failure, as well as reducing operating costs. Due to harsh marine environments, achieving reliable operational health and performance for rotor and blade components is crucial. Fault diagnoses and maintenance operations become more challenging in the sea, leading to performance degradation, failure, or breakdown of the entire tidal energy system if unattended. Therefore, the evidence-based structure health monitoring (SHM) system that will be developed in the research project could effectively avoid unplanned shutdowns or catastrophic failure. For example, by diagnosing early signs of faults/damages in the tidal turbines allows pre-emptive maintenance scheduling. For the development of the SHM system, along with the statistical approaches, the explainable artificial intelligence (XAI) based methods will be helpful to detect the deterioration of the tidal components and allow the observer to comprehend and trust the decision. The evidence based SHM system will increase reliability and improve performance over the entire tidal turbine life in complex environmental tidal conditions.