滚动轴承故障分形诊断中的信号处理

Signal Processing in Fractal Fault Diagnosis of Rolling Bearings

故障轴承振动信号具有非平稳和非线性的特征,因此可用关联维数来刻划其真实特性,但实测信号中的噪声影响了诊断的准确性。本文提出利用改进的小波包分解和重构算法对滚动轴承振动信号进行降噪处理。实验表明,该算法可有效提高轴承故障分形诊断的准确性和可靠性。

The correlation dimension can describe the real condition of the instable and nonlinear vibration signals of faulty bearings. However, the noise in the measured signals may affect the accuracy of the diagnosis. The ameliorated wavelet packet analysis and reconstruction algorithm were used to denoise the signals, which proves to be effective in improving the accuracy and reliability of the fractal fault diagnosis of bearings.