To solve the problems of weak fault signals in the early stages of rolling bearing faults and difficulty in identifying fault characteristics,a method for early weak fault diagnosis of rolling bearings based on multi-algorithm fusion is proposed.First,the multi-wavelet adaptive threshold method is used to denoise the bearing fault signal.Second,the fast spectral kurtosis is used to extract the bearing impact signal with the largest spectral kurtosis from the denoised signal.Then,an ensemble empirical mode decomposition(EEMD)dissolved the residual signal.The first three eigenmodal components(IMF)with the largest kurtosis are selected based on the kurtosis criterion and superimposed with the bearing impact signal to obtain the reconstructed signal.Finally,fast spectral correlation is performed on the reconstructed signal to enhance the periodic component and realize early weak fault diagnosis of the bearing.The simulation experiments show that the multi-algorithm fusion method for early weak fault diagnosis of rolling bearings can effectively reduce the interference of noise components,enhance weak fault characteristics,and realize early weak fault diagnosis of rolling bearings.
关键词
滚动轴承/多小波/快速谱峭度/快速谱相关/特征提取
Key words
rolling bearing/multi-wavelet/fast spectral kurtosis/fast spectral correlation/feature extraction
维普
万方数据