摘要
局部特征尺度分解(Local characteristic scale decomposition,LCD)是一种新的自适应时频分析方法,该算法可以自适应地将一个复杂信号分解为若干个ISC(Intrinsic scale component,ISC)分量之和;将LCD算法同EMD(Empirical mode decomposition),LMD(Local mean decomposition,LMD)算法进行对比分析,仿真信号的分解结果表明:三种方法都可以有效地对信号进行分解,但LCD算法在计算效率和抑制端点效应等方面要优于EMD,LMD算法;将LCD算法和zoom-FFT分析应用于齿轮故障诊断中,分析结果表明了这种相结合的方法,可以较好地提取故障特征,显示了该方法的有效性。
Local feature scale decomposition （LCD） is a new adaptive time--frequency analysis algorithm, which can adaptively decom- pose a complex signal into a number of ISC （Intrinsic scale component, ISC） components. For comparative analysis of the LCD, EMD （Em- pirical mode decomposition）, LMD （Local mean decomposition, LMD） algorithm, the decomposition results of the simulate signal show that： the three methods can effectively decompose the signal, but LCD is superior to EMD and LMD in computational efficiency and restric- tion of end effects. The LCD and zoom--FFT algorithm was used for fault diagnosis of gear, the analysis results indicate that the integrated algorithm can extract the fault feature effectively, showing the effectiveness of this method.
出处
《计算机测量与控制》
北大核心
2014年第2期352-354,共3页
Computer Measurement &Control
基金
国家自然科学基金(51275422)
关键词
局部特征尺度
经验模态分解EMD
时频分析
故障诊断
齿轮
local characteristic scale decomposition
empirical mode decomposition
time--frequency analysis
fault diagnosis
gear
