旋转机械故障诊断的时频分析方法比较  

Time-frequency Analysis and Comparison for Fault Diagnosis of Machinery

作　　者：覃慧[1] 南江红[1] 

机构地区：[1]西北工业大学机电学院,西安710072

出　　处：《科学技术与工程》2011年第9期1991-1994,共4页Science Technology and Engineering

摘　　要：旋转机械故障振动信号具有非平稳的特性,传统的傅里叶变换方法难以满足分析非平稳信号的要求,时频分析是非平稳信号分析的有力工具。以滚动轴承外圈故障信号为研究对象,对比分析研究了几种时频分析方法如STFT、Wigner-Ville分布、小波变换和Hilbert-Huang变换。结果表明:Hilbert-Huang变换能够更加清晰、准确给出故障信号时频分布,分析效果优于STFT、Wigner-Ville分布、小波分析(WT)。Aim at non-liner and non-stationary characteristic of rotating machinery vibration signals,the traditional signal analysis method（FFT） can not be satisfactory for these requirements.While time-frequency method is a useful tool.The signals of fault of rolling as the object and study sever time-frequency methods such as STFT,CWD、WT and HHT are taken.Comparing between these methods,it was demonstrated that STFT,CWD and WT method have limitations of themselves.While,HHT could detect components with lower energy,and gives a true distinct time-frequency distribution.Therefore,HHT is a effective tool to dedect early faults of rotating machinery,and it provides new means for state detection and fault diagnosis.

关 键 词：故障诊断 STFT WIGNER-VILLE分布 HHT 

分 类 号：TP206[自动化与计算机技术—检测技术与自动化装置]