全矢积频谱在滚动轴承故障诊断中的应用

摘要

局部均值分解(LMD)可将采集的时域信号分解为多个单分量信号(PF),全矢谱(FVS)技术可将双通道信息相互融合,防止单通道信息不完整.在此基础上,借鉴边际谱的思想,提出了一种新的解决方式—积频谱(FAS):采集滚动轴承的同源双通道振动信号,用LMD对同源双通道的振动信号进行处理,得到双通道各个分量的瞬时幅值和调频信号,并对调频信号进行计算得到各个分量的瞬时频率,由此可求出各通道LMD的时频分布;对时频分布进行频率上的积分后,再通过傅立叶变换求出各通道的积频谱;并通过信息融合,将得到的全矢积频谱和单通道积频谱进行对比.选择有外圈故障的滚动轴承进行试验,试验结果表明,该方法是有效的.%The local mean decomposition(LMD)can make a time signal decomposed into series of single-component signals (product function, PF), the full vector spectru m(FVS)can fuse two channels and solve the problem about incomplete information of single signal channel.On the basis of the above, drawing lessons from the thought of marginal spectrum, this paper proposes a new method:frequency accumulation spectrum(FAS)method:Firstly,two-channel vibration signals of the rolling bearings were collected and processed by LMD for the instantaneous amplitudes and frequency modulated signals,and the frequency modulated signals were calculated for instantaneous frequencies, so the time-frequency distribution of each channel based on LMD were obtained. Then the frequency accumulation spectrum of each channel can be calculated after integration on frequency and FT. By using information fusion technology of FVS gets the full vector frequency accumulation spectrum(FV-FAS),and then compares theFV-FAS with the single channel FAS.Through the fault diagnosisexperiment of rolling bearing outer ring,the results verified the validity and accuracy of this method(FV-FAS).