基于ITD和WVD的旋转机械故障诊断方法

Rotating machinery fault diagnosis method based on ITD and Wigner-Ville distribution

针对Wigner-Ville分布(WVD)分析多分量旋转机械故障振动信号存在交叉项干扰的问题,提出一种基于本征时间尺度分解(ITD)和WVD的旋转机械故障诊断方法。首先利用ITD将原始振动信号分解为若干个合理旋转(PR)分量,然后运用相关系数原则剔除其中的伪分量,再对每个真实的PR分量进行WVD分析,最后将分析结果重构并提取原信号的时频分布特征。仿真分析结果表明:该方法保留了ITD和WVD的优点,同时能有效抑制WVD的交叉项干扰,分析效果优于平滑伪Wigner-Ville分布(PWVD)。同时该文给出转子油膜涡动的故障诊断实例,验证了该方法的工程实用性。

A rotating machinery fault diagnosis method based on intrinsic time-scale decomposition(ITD) and Wigner-Ville distribution(WVD) was proposed for the cross-term interference problem of WVD in analyzing multi-component rotating machinery vibration signals. Firstly, original vibration signal was decomposed into some proper rotation(PR) components. Secondly, false PR components were deleted according to the principle of the correlation coefficient. Thirdly,WignerVille distributions of true components were computed. Finally,all the analysis results of every true component were added to extract the time-frequency distribution characteristics of the original signal. An analysis on simulation signals shows that this method preserves the advantages of ITD and WVD, the cross-term in WVD can be effectively suppressed, and the result is better than the smoothed pseudo Wigner-Ville distribution(PWVD). A rotor oil whirl instance was presented to validate the practicability of this approach.