基于HHT及共振解调方法的动车组走行部轴箱轴承故障诊断算法

Fault diagnosis algorithm for the axle box bearing of walking unit in EMU based on HHT and resonance demodulation method

针对动车组走行部轴箱振动信号非平稳、非线性且背景噪声频率分布范围广的复杂特性,设计了改进的希尔伯特-黄变换算法(Hilbert-Huang Transform,HHT),综合利用了HHT方法有效平稳化信号及共振解调方法有效提高信噪比的优点.通过分析动车组运行过程中轴箱轴承背景噪声成分,进行了轴箱轴承故障信号的仿真.对仿真信号分别使用HHT方法,共振解调方法及本文所设计方法进行故障分析.结果表明:相比HHT方法与共振解调方法,本文所设计方法可有效提取故障信息并分析故障类型.适用于分析高速动车组走行部特有的振动信号特征.为我国高速动车组故障诊断提供了新思路.

Aming at the feature of the vibration signal of axle box of walking part in EMU（Electric Multiple Unit）,which is unsmoothing,non-linear and the background noise with widefrequency range,this paper designs an improved algorithm based on HHT.The algorithm comprehensively combines the advances of HHT and resonance demodulation,of which HHT is able to smooth the signal effectively and resonance demodulation method could enhance the signal and noise radio.The background noise component of the bearing during the operation process of EMU is analyzed to conduct a simulation of fault signal.After this,the fault analysis is compared respectively using the HHT,resonance demodulation and the improve designed algorithm.The results show that HHT and resonance demodulation cannot extract fault information effectively.However,the designed algorithm is able to extract fault information effectively which can also analyze the fault type.This algorithm is appropriate to analyze the vibration signal char acteristicsof the walking part of EMU and offer a new thinking for fault diagnosis of EMU.