Data-driven train bearing fault diagnosis method

Abstract

The invention discloses a data-driven train bearing fault diagnosis method. The method comprises the following steps: acquiring an acoustic signal of a to-be-detected bearing; determining a frequency domain accumulated value according to the acoustic signal; determining a normalized logarithm Bayesian factor according to the frequency domain accumulated value based on a nonparametric probability regression model; wherein the nonparametric probability regression model is established based on a correlation vector machine and acoustic signals of the lossless bearing; and determining a diagnosis result of the bearing to be detected according to the normalized logarithm Bayesian factor. According to the diagnosis method, the data-driven sparse model about the accumulated frequency domain value is established through the relevance vector machine, the model is simple, calculation is rapid, and the health state of the bearing can be monitored in real time. The frequency domain accumulated value is put forward for modeling for the first time, the mechanism is simple, and the effect is direct and obvious. 本发明公开了一种数据驱动的列车轴承故障诊断方法，包括步骤：获取待测轴承的声学信号；根据所述声学信号，确定频域累积值；基于非参数概率回归模型，根据所述频域累积值，确定归一化对数贝叶斯因子；其中，所述非参数概率回归模型是基于相关向量机和无损轴承的声学信号建立的；根据所述归一化对数贝叶斯因子，确定所述待测轴承的诊断结果。本发明诊断方法通过相关向量机建立关于累积频域值的数据驱动稀疏模型，模型简单，计算快速，可以实时监控轴承健康状态。频域累积值被首次提出用于建模，机制简单，效果直接明显。