摩擦振动信号谐波小波包特征提取

在销-盘摩擦磨损试验机上进行了船用柴油机缸套-活塞环摩擦配副摩擦磨损试验,应用谐波小波包变换分析了不同磨损状态下的摩擦振动信号.结果表明:谐波小波包变换可提取微弱的摩擦振动特征.当磨损机制为轻微的黏着磨损时,摩擦振动信号的频谱以边频为主,没有明显的固有频率,平均幅值小;当磨损机制以磨粒磨损为主时,摩擦振动信号的频谱出现较大峰值的固有频率,但仍存在大量的边频,平均幅值变大;当磨损机制以疲劳磨损为主时,摩擦振动信号频谱的固有频率更加明显,边频显著减弱,平均幅值最大.因此,谐波小波包变换可实现微弱摩擦振动信号的特征提取.

Characteristics Extraction of Friction Vibration Signal Using Harmonic Wavelet Packet Transforms

The friction and wear tests of a piston ring against a cylinder from marine diesel engine were conducted on a pin-on-disc tester. The friction vibration signals under different wear condition were studied making use of the hamonic wavelet packet transform. It was showed that the weak friction vibration signal can be extracted using the hamonic wavelet packet transform. When the wear mechanism is slightly adhesion wear, the side-frequencies dominate the frequency spectrum of friction vibration signals, the natural frequencies don't appear in the frequency spectrum and the average amplitude of the signals is small. When the wear mechanism is abrasive wear, the natural frequencies appear in the frequency spectrum of friction vibration signals, but a large number of side-frequencies still exist and the average amplitude of the signals increases. When the wear mechanism is fatigue wear, the natural frequencies dominate the frequency spectrum of friction vibration signals, the side-frequencies significantly reduce and the average amplitude of the signals is big. Therefore The characteristic extraction of that the weak friction vibration signal can be carried out by the hamonic wavelet packet transform.