基于小波包和RBF神经网络的轨道车辆滚动轴承故障诊断

摘要

In this paper, a method combined wavelet packet with RBF Neural Network was proposed for the fault diagnosis of the rolling bearing of railway vehicles. First, wavelet denoising was performed for the collected vibration signal. And then, energy characteristic vectors of the fault signals were extracted by wavelet packet decomposing. At last, the extracted energy characteristic vectors were used to train the RBF Neural Network. The diagnostic results showed that the proposed method could diagnose fault types of the rolling bearings of railway vehicle precisely. The proposed method had certain practical application value.%针对轨道车辆的滚动轴承故障诊断问题,提出了一种小波包与RBF神经网络相结合的故障诊断方法.首先对采集到的振动数据进行小波消噪,然后利用小波包分解提取故障信号的能量特征向量,最后利用提取的能量特征训练RBF神经网络,进行故障诊断.诊断结果表明,基于小波包和RBF神经网络的轨道车辆滚动轴承故障诊断方法能够较好的诊断出轨道车辆的轴承故障类型,具有一定的实际应用价值.