一种适用于发动机振动信号的时频分析方法

摘要

本文中对一种新的时频分析方法——同步压缩小波变换进行研究.首先,为检验其对发动机信号的适用性,建立了多分量﹑调幅调频仿真信号,采用同步压缩小波变换对其进行分析.接着将其与其它时频分析方法在时频分辨率﹑信号分解和重构能力方面进行比较;最后以某一发动机为例,分析其瞬变工况下的振动信号,揭示连杆轴承磨损信号变化规律并提取故障特征.结果表明:同步压缩小波变换是一种适用于发动机状态监测与故障诊断的时频分析方法.%A new time-frequency analysis method-synchrosqueezing wavelet transform ( SWT) is studied in this paper. Firstly for checking its suitability for engine vibration signals, multi-component, AM-FM simulation sig-nals are set up and analyzed by SWT. Then SWT is compared with other time-frequency analysis methods in terms of resolution and capabilities of signal decomposition and reconstruction. Finally, with an engine as an example, its vi-bration signals in transient conditions are analyzed and the changing rule of connecting rod bearing wear signals is revealed with fault features extracted. The results show that SWT is an appropriate time-frequency analysis method suitable for engine state monitoring and fault diagnosis.