Machine Learning-based OWC Diagnosis Using Real Measured Data from Wave Power Plants

The present manuscript introduces a classification for power take-off (PTO) diagnosis in Wave Energy Converter (WEC) farms. The suggested strategy has been tested on the Mutriku Multiple Oscillating Water Column-based wave power plant in order to reduce the Levelized Cost of Energy (LCoE) by impleme...

ver descrição completa

Detalhes bibliográficos
Autores: M'Zoughi, Fares, Lekube Garagarza, Jon, Garrido Hernández, Izaskun, Garrido Hernández, Aitor Josu
Formato: artículo
Fecha de publicación:2024
País:España
Recursos:Universidad del País Vasco
Repositorio:Addi. Archivo Digital para la Docencia y la Investigación
OAI Identifier:oai:addi.ehu.eus:10810/73741
Acesso em linha:http://hdl.handle.net/10810/73741
Access Level:acceso abierto
Palavra-chave:classification
fault diagnosis
LDA
machine jearning
oscillating water column
power-take off
SVM
wave energy
Descrição
Resumo:The present manuscript introduces a classification for power take-off (PTO) diagnosis in Wave Energy Converter (WEC) farms. The suggested strategy has been tested on the Mutriku Multiple Oscillating Water Column-based wave power plant in order to reduce the Levelized Cost of Energy (LCoE) by implementing predictive maintenance strategies. This has been achieved by employing Linear Discriminant Analysis (LDA) to determine the furthermost relevant features from the measured data. Then the Support Vector Machine (SVM) has been implemented as a classification technique to classify the state of the OWC system.