中文摘要：

提出一种基于谱峭度和Morlet小波的滚动轴承微弱故障诊断新方法.该方法利用Morlet小波的滤波特性,基于谱峭度表征的Morlet小波系数最大化原则,采用频带平移及外扩的方法自适应地在全频范围内定位由冲击信号激起的各个共振频带,再利用谱峭度和带宽这两个参数有选择地利用一个或若干个共振带进行轴承微弱故障诊断.与原有的谱峭度快速算法相比,该方法能有效避免一个共振带被人为分割为多个频带以及将非共振频带并入共振频带中的现象,从而提高了轴承故障诊断效果.另外,该方法避免了传统方法只利用一个共振带而丢弃其他频带的做法,可以利用找到的不同共振频带的轴承信息进行微弱故障特征增强,以提高微弱故障诊断性能.滚动轴承微弱故障实验证明了新方法的有效性.

英文摘要：

A new weak fault diagnosis method for rolling element bearings based on Morlet wavelet and spectral kurtosis is proposed.With the band-pass character of Morlet wavelet and the principle of the spectral kurtosis maximization,this method searches all the possible resonance frequency bands adaptively by shifting and expanding predetermined Morlet wavelets.One or several resonance frequency bands are chosen for rolling element bearing weak fault diagnosis according to the spectral kurtosis and the bandwidth.Compared with the Fast Kurtogram,the new method can avoid a resonance frequency bands being divided into several sections artificially,or useless frequencies being included in the resonance frequency band.On the other hand,the new method improves the weak fault diagnosis ability by utilizing several resonance frequency bands to enhance fault characters.The effectiveness of the proposed method is proved by the rolling element bearing weak fault experiments.