Wind Turbine Engineering Management: A Real Case Study with Acoustic Monitoring

Wind energy has emerged as one of the most relevant renewable energy sources, requiring novel condition monitoring systems to reach competitiveness and ensure the reliability and maintainability of wind turbines. This paper introduces an innovative non-destructive testing system designed for the acoustic analysis of wind turbine rotatory components, capturing the acoustic signals from the nacelle using an acoustic condition monitoring system. A signal processing procedure employing wavelet transforms to filter and analyse the energy samples is presented, to detect patterns associated with different scenarios. The implementation of different wavelet families together with novel analytic techniques provided high accuracy in pattern classification results. A real case study is presented to test the methodology, concluding that acoustic monitoring is a suitable and reliable technique for fault detection in wind turbines.