Fault Recognition of Wind Turbine Using EMD Analysis and FFT Classification

This paper employs empirical mode decomposition (EMD) and fast Fourier transform (FFT) to analyze the oil-leakage fault signal of the gearbox of wind turbines. K-nearest neighbors (KNN) is used on automatic fault recognition. First, both normal and faulty oil-leakage gearboxes are considered. Second, EMD is applied on analyzing the intrinsic mode function (IMF) of the current signals, and FFT is used to get the IMF spectrum. Finally, the features of the spectrum are extracted, and KNN is used on fault recognition of wind turbine gearboxes. The result indicates that it can effectively recognize the oil-leakage fault of gearboxes.