Building an Air Turbine Conditional Anomaly Detection Approach for Wave Power Plants

Abstract

The Mutriku Wave Power Plant (WPP) is a wave energy conversion plant based on the oscillating water column technology (OWC). The energy production and the health state of the plant are directly dependent on the sea-state conditions along with component-specific operation efficiency and failure modes. In this context, this paper presents a preliminary air turbine conditional anomaly detection (CAD) approach for condition monitoring of the Mutriku WPP. The proposed approach is developed based on an ensemble of Gaussian Mixture models, where each anomaly detection model learns the expected air turbine operation conditioned on specific seastates information. Early results show that the integration of sea-states in the anomaly detection learning process improves the discrimination capability of the anomaly detection model.