摘要
为了消除机械振动信号中趋势项,提出一种基于局部均值分解(LMD)的趋势项提取新方法.考虑到机械振动信号中夹杂的趋势项特点,其残余量并不会对趋势项产生过大的影响,所以可将LMD的最后一个频率分量作为待提取的趋势项.为了验证该方法的可靠性,通过数值模拟的方法将线性、多项式和指数趋势项加载到待消除的仿真信号和实测滚动轴承振动信号中,并将笔者所提的LMD方法与常规的最小二乘拟合法、小波分析和EMD方法分别用于趋势项的提取.数值模拟试验和实测数据分析表明,该方法能够较准确地提取机械振动信号的趋势项,与其他方法相比具有一定的优势.
A novel method of trend elimination for mechanical vibration signal by local mean decomposition is proposed in this paper.LMD is a decreasing decomposition process from high-frequency component to low-frequency component step by step,and both in amplitude and frequence of the residual component which is very weak.In view of the low frequency for the trend of mechanical vibration signals,and the residual component from LMD has little impact on the trend.Therefore,the last PF component could be the trend for mechanical vibration signals.In order to validate the correctness of proposed method in this paper,by using the method of numerical simulation to build the common linear trends,polynomial and exponential trend for loading on the simulation signal and the measured vibration signals of rolling bearings.The proposed LMD method,the conventional least square fitting method,the wavelet analysis and the EMD method can be used to extract the trends respectively.Numerical simulation test and the measured data processing results show that this method can accurately extract the trend of the mechanical vibration signal,compared with the other methods to extract the trend,and LMD has a certain advantage.
出处
《郑州大学学报（工学版）》
CAS
北大核心
2014年第5期100-104,共5页
Journal of Zhengzhou University（Engineering Science）
基金
河南省科技攻关资助项目(142102210048)
关键词
机械振动信号
局部均值分解
趋势项
包络信号
纯调频信号
mechanical vibration signal
local mean decomposition
trend
envelope signal
purely frequency
