基于多小波和改进双树复小波的滚动轴承故障诊断方法  

Rolling bearing fault diagnosis method based on multi-wavelet and improved double-tree complex wavelet

作　　者：郭建超 卢岩(指导) 陈秦箫 袁驰 GUO Jianchao;LU Yan;CHEN Qinxiao;YUAN Chi(School of Electrical Engineering,Shanghai Dianji University,Shanghai 201306,China)

机构地区：[1]上海电机学院电气学院,上海201306

出　　处：《上海电机学院学报》2023年第2期90-96,共7页Journal of Shanghai Dianji University

摘　　要：当旋转机械中发生局部缺陷时,其振动信号往往由周期性冲击分量及其他分量构成,其中冲击分量反映滚动轴承的状况。由于强烈的背景噪声以及信号的耦合作用,滚动轴承的故障特征频率往往被模糊。针对该问题提出了一种基于多小波和改进双树复小波的滚动轴承故障诊断方法。首先,对采集到的振动信号进行多小波相邻系数自适应阈值降噪处理;然后,对降噪后的信号进行双树复小波分解,利用鲁棒局部均值分解获取各小波分量的主频率分量;最后,利用基尼系数选择出最佳子带并进行包络谱分析,实现轴承的故障诊断。通过轴承故障仿真和实测数据的分析对比,证明了该方法可有效辨别出滚动轴承的典型故障,提高滚动轴承故障的诊断效果,同时与快速谱峭度进行对比,表明了该方法的优越性。When there are local defects in rotating machinery,the vibration signal is usually composed of periodic impact components and other components,and the impact components reflect the condition of rolling bearings.Due to the strong background noise and the signal coupling,the fault characteristic frequency of rolling bearings is often blurred.To solve this problem,a rolling bearing fault diagnosis method based on multi-wavelet and improved double tree complex wavelet is proposed.Firstly,an adaptive threshold noise reduction based on a multi-wavelet adjacent coefficient is performed on the collected vibration signal.Then,the double-tree complex wavelet decomposition is performed on the noise-reduced signal,and the robust local mean decomposition is used to obtain the main frequency components of each wavelet component.Finally,the Gini coefficient is used to select the optimal sub-band and the envelope spectrum analysis is performed on it to realize the fault diagnosis of the bearings.The analysis and comparison of bearing fault simulation and measured data show that the typical faults of rolling bearings can be effectively identified and the diagnosis effect of rolling bearing faults is improved by the method.At the same time,the superiority of the method is proved by the comparison with the fast spectral kurtosis.

关 键 词：多小波相邻系数自适应阈值降噪 双树复小波变换 鲁棒局部均值分解 基尼系数 

分 类 号：TH165.3[机械工程—机械制造及自动化]