Nondestructive structural health monitoring of composite wind turbine blades using acoustic emission

Abstract

Structural Health Monitoring (SHM) is critical to guarantee that composite wind turbine blades (WTBs) operate efficiently and reliably. Effective SHM can diminish downtime, lower maintenance costs, and increase energy production, while providing industrial systems with improved safety. This study introduces a novel, simplified approach to nondestructive SHM for glass-fibre reinforced composite (GFRC) blades, utilizing Acoustic Emissions (AE). To identify damage sources, AE signals generated by laboratory testing of damaged GFRC blades are captured and processed into Red, Green and Blue spectrograms, allowing for representing more characteristics of the raw data. A custom-designed machine learning model is then used to extract features from these spectrograms, enabling damage detection. This method provides a practical SHM solution for WTBs in operation, incorporating a sensor network for real-time monitoring.