Classification-based OWC Diagnosis Using Real Measured Data From Mutriku WavePower Plant
- Fares M’zoughi 1
- Jon Lekube 2
- Izaskun Garrido 1
- Aitor J. Garrido 1
- 1 Automatic Control Group - ACG, Department of Automatic Control and Systems Engineering, Faculty of Engineering of Bilbao - EIB/BIE, University of the Basque Country - UPV/EHU
- 2 Biscay Marine Energy Platform, BiMEP
- Aitor J. Garrido Garrido (coord.)
- Matilde Santos Peñas (coord.)
- Fares Mzoughi (coord.)
- Ahmad, Irfan (coord.)
- Garrido Hernandez, Izaskun (coord.)
Editorial: Servicio Editorial = Argitalpen Zerbitzua ; Universidad del País Vasco = Euskal Herriko Unibertsitatea
ISBN: 978-84-1319-526-1
Año de publicación: 2023
Páginas: 83-88
Congreso: Jornada Internacional de Energía Eólica y Marina (4. 2022. null)
Tipo: Aportación congreso
Resumen
This paper presents a classification-based PTO diagnosis for wave energy converter farms. The proposed technique is applied to the study case of Mutriku MOWC wave power plant in order to reduce the LCoE by implementing predictive maintenance strategies. To do so, the work adopted the well-known extraction methods, Principal Component Analysis (PCA) to select the most relevant features for OWC diagnosis. In addition, the Support Vector Machine (SVM) classification method has been used to classify the wave power plant data. The obtained data show that the SVM classification method allows achieving a successful performance with a high degree of accuracy when using PCA extraction method. The implementation of a predictive maintenance over the Mutriku wave power plant using the developed classification-based OWC diagnosis indicate that, with this predictive maintenance, the OpEx can be reduced up to 17%, downtime can be reduced to 20% and plant availability can be increased up to 81%. These indicators lead to a reduction of the LCoE by 23%, which represents about a quarter of the total costs.