基于EMD和AFD的齿轮断齿诊断技术研究

Research on Broken Teeth Diagnosis Technology Based on EMD andAFD

断齿作为齿轮失效的形式之一,对其进行故障诊断具有重要意义。传统的诊断方法如自适应滤波方法对于非平稳振动信号的分析效果不太明显。EMD(Empirical Mode Decomposition,简称EMD)方法是把复杂的振动信号分解为有限个本征模函数(Intrinsic Mode Function,简称IMF)之和,得到的IMF包含真实的物理信息,而且都是平稳的。这种基于信号的局部特征时间尺度分解的方法非常适用于非线性和非平稳过程的分析,该方法能够实现通过时域的分析,得到故障特征信号。本文提出基于EMD和自适应滤波分解(Adaptive Filtering Decompositon,简称AFD)的方法,利用边际谱分析齿轮故障特征。实验结果表明,该方法能够有效、准确地提取齿轮断齿的故障特征。

As one of the failure forms of gear, broken teeth on the fault diagnosis is of great significance. Traditional diagnostic methods such as adaptive filtering decompositon tbr non-stationary vibration signal＇s analysis effcction is not obvious. The Empirical Mode Decomposition is a method that decomposes the complex vibration signals into the sum of a finite number oflntrinsic Mode Function, the IMF contains the actual physical information, and is smooth. This method which based on the local characteristic time scale of signal decomposition is very suitable for nonlinear and non-stationary process analysis, and it can realize getting fault characteristic signal through the analysis of time domain. This paper proposes a method, based on the EMD and Adaptive Filtering （AFD）, Decomposition uses marginal spectrum analysis gear fault characteristics. The experimental results show that the method can effectively and accurately extract the fault characteristics of gear tooth broken.