Enhancing Wind Turbine Generator Diagnostics via Wavelet Transform: A Multi-Algorithm Approach for Fault Detection

Abstract

This study presents an advanced diagnostic method for wind turbine generators based on the Wavelet Transform (WT), aimed at enhancing the accuracy of fault detection and localization.
By analyzing the stator and rotor currents, the method leverages WT's capacity to process nonstationary and transient signals, which is particularly effective for capturing dynamic anomalies in wind energy systems. Three distinct algorithms—Morlet, Gabor, and Wigner-Ville—were implemented to classify faults in both stator and rotor circuits. These approaches collectively form a robust framework suitable for real-time condition monitoring. The diagnostic method significantly improves early fault identification, minimizes operational downtime through precise localization, and enhances the reliability of wind energy generation systems. Moreover, it enables the optimization of maintenance strategies by facilitating the targeting of specific fault types. Experimental results confirm the effectiveness of WT in converting raw current data into diagnostically relevant insights, thereby contributing to the development of more efficient and resilient wind energy infrastructures.