基于分形理论、熵理论及小波变换方法,提出用关联维数和小波能量谱熵对转子系统不同碰摩程度进行量化检测的方法。首先,利用转子碰摩模型仿真了不同碰摩程度下的振动信号,并计算了信号的关联维数和小波能量谱熵,研究了关联维数和小波能量谱熵随转子不同碰摩程度的变化规律,发现这两种特征量与转子碰摩程度之间存在良好的相关性,表明其可以作为转子碰摩程度量化检测的重要特征量。最后,利用ZT-3多功能转子实验器,模拟了不同碰摩状态的振动信号,对该方法进行实验验证。结果表明:关联维数和小波能量谱熵对转子碰摩严重程度量化检测的有效性。
Based on the fractal and entropy theory and the wavelet transform method,a quantitative detection method for rubbing intensity of rotor system by use of correlation dimension and wavelet energy spectrum entropy is proposed in this paper.First,we calculate the correlation dimension and wavelet energy spectrum entropy of the vibration signals which are simulated by use of the rub-impact rotor model with different rubbing intensity.Then,we study the change law of the correlation dimension and wavelet energy spectrum entropy with different rubbing intensity.It is found that the two features are well correlated with the rubbing intensity of the rotor,which indicates that the two features can be used as the important characteristic quantities for quantitative detection of rubbing intensity.Finally,we calculate the correlation dimension and wavelet energy spectrum entropy of the vibration signals which are simulated by use of the ZT-3 multi-function rotor test rig with different rubbing intensity.Results verify the above method,and show that it is effective to quantitatively detect the rubbing intensity of the rotor system by use of the correlation dimension and wavelet energy spectrum entropy.