基于倒谱预白化的轨道车辆轴箱轴承故障诊断

Fault diagnostics of axle-box bearing in railway vehicles based on cepstrum editing procedure

摘要:

轴箱轴承作为轨道车辆的关键旋转部件之一，起着承载、减磨、降阻的作用。基于振动信号的轴承监测方法，将成为轨道车辆轴承状态在线监测的发展方向。在信号预白化的理论基础上，提出了基于倒谱编辑预白化的轴承故障诊断方法，该方法使用实倒谱定位和编辑原始信号的对数幅值，去除不想要的离散频率分量，然后结合原始相位信号和编辑后的幅值重构时域信号，再对时域白化信号进行频谱分析。试验结果表明，倒谱预白化与传统包络谱解调分析都能有效检测轴箱轴承故障，但倒谱预白化不需要进行滤波频带选择，具有简单实用的优点，验证了该方法在轴承严重故障和微弱故障诊断方面的有效性。

Abstract:

Axle-box bearing is one of the key rotating parts in rail vehicles, which sustains load as well as reduce friction. Currently, the on-board method is based on temperature monitoring or vibration monitoring. Besides, the wayside method is based on acoustic monitoring. The vibration based method will be the main technique for condition monitoring of axle-box bearing, due to the abundant algorithm study and engineering application. This work integrates the cepstrum editing procedure and the envelope analysis to diagnose the axle-box bearing fault qualitatively. The effectiveness of the improved method to identify both the weak fault and the serious fault is tested on the rolling test rig.