多尺度排列熵及其在滚动轴承故障诊断中的应用

Multi-scale Permutation Entropy and Its Applications to Rolling Bearing Fault Diagnosis

引入多尺度排列熵(MPE)的概念,用来检测振动信号不同尺度下的动力学突变行为,并将其应用于机械故障诊断中滚动轴承故障特征的提取,结合支持向量机(SVM),提出了一种基于MPE和SVM的滚动轴承故障诊断方法,将新提出的滚动轴承故障诊断方法应用于实验数据分析,并通过与BP神经网络对比,结果表明,该方法能够有效地提取故障特征,实现故障类型的诊断。

A definition of MPE was presented to extract the fault characteristics of dynamics chan-ges from bearing vibration signals. And in combination with SVM, a bearing fault diagnosis approach was put forward based on MPE and SVM. Firstly the algorithms of PE and MPE were introduced. Then experimental data were used to demonstrate the validity of the approach. Also for comparision with SVM, the BP neural network was used and the analysis results indicate that the proposed ap-proach can extract the fault feature and identify the fault categories effectively.