基于频率截止的EEMD方法研究

An improved EEMD method based on cut-off frequency

为解决总体平均经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)中虚假IMF分量过多问题,提出了一种基于频率截止的EEMD方法。该方法采用一种新的IMF筛分终止条件——以信号自身的最小频率为EMD分解IMF分量的截止频率;然后将基于频率截止的IMF筛分终止条件引入EEMD分解。通过仿真和实测信号分析,并与EMD、EEMD分解结果比较得到,运用频率截止的EEMD方法不仅有效减少了虚假IMF分量的产生,使得分解的目的性更加明确,而且保证了EEMD分解出的IMF分量的完备性,更好地抑制了模态混叠现象。

In order to solve the problem of excessive false IMF components in ensemble empirical mode decomposition（EEMD）,an improved EEMD method based on cut-off frequency was proposed here.This method adopted a new screening termination condition of IMF,the minimum frequency of a signal itself was taken as the cut-off frequency when using EMD.Then,the screening termination condition of IMF based on cut-off frequency was introduced into the EEMD.Comparing the results of EMD and EEMD with those of the improved EEMD,it was shown that the false IMF components are reduced more effectively and the mode mixing phenomenon is suppressed with the improved EEMD method based on cut-off frequency in analyses of simulated and actually measured signals;the IMF components are complete and the purposiveness of decomposition is more clear with this proposed method.