关键词: 齿轮箱, 互补集合经验模态分解, 细化快速傅里叶变换, 小波包, 特征提取, 故障诊断

Abstract: Fault diagnosis of wind turbine gearbox based on signal processing is an important research direction in the field of wind power generation. A method based on Zoom-FFT-CEEMD and wavelet packet noise reduction was proposed for the problem of wind turbine gearbox fault feature extraction. The RMS trend analysis was done on the vibration acceleration signals of each measurement point collected at the vibration measurement points of the wind turbine gearbox to find out the measurement points and time periods where the RMS trend rises significantly. The wavelet packet noise reduction technique was used to reduce the noise of the vibration signals at the measurement point, followed by multi-scale analysis of the vibration signals with the components obtained from the complementary ensemble empirical modal decomposition ( CEEMD) decomposition, and the Zoom algorithm was used to locally amplify the vibration signals of the gearboxes in order to highlight the fault signals. The amplified signal was analysed spectrally using fast fourier transform ( FFT) to improve the accuracy of fault feature extraction. The experimental results show that compared with the traditional spectrum analysis method, the method could effectively extract the fault features of the wind turbine gearbox with high accuracy and stability, providing an effective means for the early fault diagnosis of wind turbine gearboxes.

Key words: gearbox,complementary ensemble empirical modal decomposition( CEEMD), Zoom-FFT,wavelet packet, feature extraction, fault diagnosis