Guided Wave-based Structural Health Monitoring of Composite Wind Turbine Blades

Monitoring of wind turbines is essential for ensuring their efficient and reliable operation, minimizing downtime and maintenance costs, and maximizing energy generation. This research introduces a novel and streamlines method for intelligent structural health monitoring (SHM) of composite wind turbine blades using ultrasonic guided wave (GW) signals and machine learning. Our methodology involves real-time monitoring of laboratory-based glass-fiber reinforced composite wind turbine blades. The GW signals obtained from experiments are processed into time-frequency spectrograms, which are then used to train, validate, and test a machine learning model designed for autonomous damage-source identification. The model uses a purposely designed machine learning model for feature extraction and classification of the blade conditions. This integration provides a robust in-service SHM system for wind turbine blades, surpassing traditional methods. The proposed approach is particularly suited for wind farm environments, with the potential for future deployment of wireless sensors. Our findings indicate that this method can significantly enhance wind turbine design, improving reliability and efficiency and contributing to eco-friendly sustainable energy generation.