基于ROC分析及样本熵的轴承故障检测

Bearing Fault Detection Based on ROC Analysis and Sample Entropy

摘要: 轴承故障检测对旋转机械的维护至关重要。尽管已有的故障诊断方法取得了长足的进步，但仍然面临着一些挑战，如缺乏足够的故障数据用于训练，对复杂的分布式数据的有效性不高，对早期故障的敏感性低，以及噪声和离群值的干扰。因此，本文提出了一种基于样本熵和ROC分析的故障检测方法，通过提取振动信号的样本熵指标，然后使用ROC分析对其进行检测。实验结果表明，本文所提方法能够以较高的准确率检测轴测故障。

Abstract: Bearing fault detection is crucial for the maintenance of rotating machinery. Although existing fault diagnosis methods have made significant progress, they still face some challenges, such as a lack of sufficient fault data for training, low effectiveness against complex distributed data, low sensitivity to early faults, and interference from noise and outliers. Therefore, this article proposes a fault detection method based on sample entropy and ROC analysis, by extracting the sample entropy index of vibration signals and then using ROC analysis to detect them. The experimental results show that the method proposed in this article can detect axial faults with high accuracy.