EMD信号分析方法边缘效应的分析

Analysis on End Effects of EMD Method

在论述了经验模式分解 ( EMD)信号分解原理的基础上 ,分析了其存在的边缘效应 ,并提出了通过添加极值点抑制边缘效应的思路和策略。然后 ,针对一个单分量非线性信号和齿轮箱振动信号 ,对比分析了添加两个极值点、一个极值点和直接以数据端点作为极值点三种方式进行 EMD分解的效果 ,所提出的预测附加极值点较好地抑制了边缘效应 ,并指出了进一步完善的方向。

The end effects of empirical mode decomposition (EMD) are discussed and the end effects are restrained by the method. The results of EMD in both signals are analyzed by three ways. The ways are carried out by adding two consecutive extremes, one extreme and taking the original end data as one extreme. Analysical results show that the end effects of EMD method by adding extreme are restrained and further research directions are pointed out.