摘要
针对在强噪声背景下轴承振动信号的非线性,非平稳性以及调制源微弱难以提取故障特征的问题,提出了一种基于小波包熵值与EMD(经验模态分解)结合的能量算子解调故障诊断方法。该方法首先对信号进行小波包熵值降噪,进而选取相关度最大的IMF(本征模态分量)进行能量算子解调,从而实现了提取该分量下的故障信号的幅值和频率信息。对机械故障振动信号进行实验分析表明,相对于普通Hilbert解调法的运算精度与运算速度满足不了诊断需求的情况下,该方法能够有效解调出故障频率信息,实现对故障类别的推断。
In view of the bearing vibration signal under strong noise background of nonlinear,non-stationary and modulation weak hard to extract fault features of the source problem,it proposes a wavelet package entropy values and the EMD（empirical mode decomposition）combined with energy operator demodulation method of fault diagnosis.This method is firstly used to the signal wavelet package entropy value of noise reduction,and then selects relevant biggest IMF（intrinsic mode components）for energy operator demodulation,realizes the extraction fault signal amplitude and frequency information under this component.Experiment on mechanical fault vibration signal analysis show that compared with the ordinary Hilbert demodulation method of computing precision and computing speed it can＇t satisfy the diagnosis demand conditions,and the method can effective demodulation failure frequency information,and realize the fault category.
出处
《机械设计与制造》
北大核心
2015年第10期217-220,共4页
Machinery Design & Manufacture
基金
内蒙古自治区高等学校科学研究项目(NJZY154)
内蒙古科技大学创新基金项目(2014QDL025)
