Faults detection in squirrel cage induction generator rotor on wind turbine: Hilbert Huang transform application

Abstract

In this study we investigate the possibility of rotor broken bar fault detection in squirrel cage induction generator of a wind turbine using spectral analysis of the stator currents. The numerical method, presented in this work, based on the Hilbert Huang transform shows the possibility of improving the detection of faults in electrical machines. Using the Hilbert Huang transform analyses the stator current signal using intrinsic mode functions, which are extracted using the process of empirical mode decomposition. However, the use of Hilbert transform based time domain approach in Hilbert Huang transform for analysis of stator current signature leads to scope for subjective error in calculation of characteristic defect frequencies of the broken rotor bars.