In response to the problem of low accuracy in variable condition fault diagnosis caused by the inability to completely eliminate the noise of high-speed bearing fault signals,this paper proposes an AMO-VMD processing method for variable condition fault data of high-speed bearings,and verifies the effectiveness of the proposed algorithm in im-proving noise reduction performance from multiple dimensions.First,the noise-reduction adjacent mode overlap(AMO)index is proposed to determine the optimal decomposition number of variational mode decomposition(VMD),which can fully leverage the advantages of VMD frequency domain analysis.Secondly,a component filtering and denoising method based on the principle of maximum envelope spectral entropy is designed to improve the accuracy of fault signal frequency domain information expression.Then,a data cleaning system for high-speed bearing fault data is constructed by combin-ing DC component separation technology and median absolute deviation(MAD)de-differentiation method.Finally,the effectiveness of the proposed algorithm is verified based on typical bearing fault simulation signals and high-speed test rig bearing fault signals,respectively.The results show that the proposed method has higher fault information discrimination ability than traditional data processing methods,and can significantly improve the accuracy of bearing fault diagnosis under variable working conditions.
关键词
数据处理/模态分解/故障检测/高速轴承/相邻模态重叠指数
Key words
data processing/mode decomposition/fault detection/high-speed bearing/adjacent mode overlap index
维普
万方数据