EMD方法和倒频谱在齿轮箱故障诊断中的应用  

Fault Diagnosis of Gearbox Based on Cepstrum and EMD Method

作　　者：刘自然[1] 熊伟[1] 颜丙生[1] 甄守乐[1] 王律强 

机构地区：[1]河南工业大学机电工程学院,郑州450007

出　　处：《组合机床与自动化加工技术》2014年第9期102-104,共3页Modular Machine Tool & Automatic Manufacturing Technique

摘　　要：将经验模态分解(Empirical Mode Decomposition,简称EMD)和倒频谱结合,利用EMD分解齿轮箱的振动信号,得到具有不同频率的固有模态函数(IMF)。从得到的基本模态函数可以发现齿轮箱振动信号具有明显的调幅特性,证明EMD方法可以有效地提取齿轮箱故障信息。将固有模态函数进行包络谱分析,运用倒频谱的方法对这些信号进行进一步提取和分离。试验表明,倒频谱和EMD方法相结合,能够准确提取振动特征。This paper combines the empirical mode decomposition（EMD）with cepstrum. EMD will make the gearbox vibration signal decomposed, and the intrinsic mode function can be various with different fre-quencies. From basic components, we can see the characteristic of amplitude modulation, indicating that EMD can be effectively applied to gearbox fault diagnosis. Then the intrinsic mode functionare applied to envelope spectrum analysis at the same time, using the cepstrum to extract and separate these signals. The experimental result shows that the combination of cepstrum analysis and EMD can effectively extract features of vibration.

关 键 词：经验模态分解 倒频谱 齿轮箱 

分 类 号：TH132.41[机械工程—机械制造及自动化] TG65[金属学及工艺—金属切削加工及机床]