摘要
为解决异步电机故障轴承振动信号易受噪音影响信噪比较小的缺点,提出了一种新的故障诊断方法。首先,采用小波分析方法对测得的原始信号进行去噪,并根据频率对原始信号进行频带划分;其次,用经验模式分解(EMD)方法对小波包分解重构得到的低频段信号进行分解,获得若干固有内在模函数(IMF);最后,采用傅里叶变换对各个IMF函数进行时频分析获得频谱图,进而提取故障频率,根据故障频率和故障类型的对应关系得出最后的诊断结果。实验表明,该方法能有效地提取出故障特征频率,方便地判断出故障类型。对比分析了傅里叶变换和小波变换与本方法的优缺点,为滚动轴承的早期故障诊断提供了一个新的思路。
A new fault diagnosing method was proposed to deal with the vibration signal of fault rolling bearings in asynchronous motor that has small signal-to-noise ratio as a reault of noise influence.First,the original signals were denoised by wavelet analysis and conducted frequency division;seconally,the low-frequency signals,which were obtained by wavelet decomposition and reconstraction,were resolved by EMD to give a number of inherent IMFs;finally,time-frequency analysis was carried out on each IMF by fourier transform to give spectrogram from which fault froquency was extracted to draw out the diagnosis result from the relation between fault frequency and fault type.This method was proved to be effective in extracting characteristic fault frequency for fault diagnosis.The advantages and weakness of this method were compared with those of fourier transform and wavelet transform.A new way for early fault diagnosis of rolling bearing was provided.
出处
《太原理工大学学报》
CAS
北大核心
2010年第2期178-182,共5页
Journal of Taiyuan University of Technology
基金
山西省科技攻关资助项目(2006031153-01)
山西省自然科学基金资助项目(2007011068)
