论文部分内容阅读
将描述混沌运动的关联维数用于转子系统振动信号分析,并针对现场实测振动信号中存在的噪声污染,引入形态滤波器对信号进行降噪处理。分析比较降噪前后转子4种状态(正常、油膜涡动、局部碰摩、全周碰摩)下信号的关联维数,结果表明利用含有噪声信号的关联维数来描述系统的特征行为是不可靠的,对现场实测信号进行降噪处理是必要的。形态滤波器具有良好的降噪效果,降噪后关联维数有效地反映和区分了转子的不同状态。将降噪后的关联维数作为转子故障诊断中识别故障、判断运行状态的特征量是可行的。
The correlation dimension of chaotic motion will be described for the vibration signal analysis of the rotor system. Aiming at the noise pollution existing in the field vibration signal, a morphological filter is introduced to denoise the signal. The correlation dimension of the signal under the four states (normal, oil film whirling, local rubbing, and rubbing) of the rotor before and after noise reduction is analyzed and compared. The results show that using the correlation dimension with noise signal to describe the characteristic behavior of the system is not Reliable, on-site measured signal noise reduction is necessary. The morphological filter has a good noise reduction effect, and the correlation dimension after noise reduction effectively reflects and differentiates the different states of the rotor. It is feasible to use the correlation dimension after noise reduction to identify faults in rotor fault diagnosis and judge the running state.