EMD中包络算法改进的研究与分析

The Research of Improved Envelope Algorithm of EMD

经验模态分解算法(EMD)是由美籍华人N.E.Huang提出的一种新的数据分析方法,已被广泛的应用于故障诊断方面的研究。但是作为新出现的信号处理,除需要进一步的理论证明外,EMD方法仍然存在着许多需要改进的地方。原始的EMD算法采用三次样条插值算法来拟合非平稳信号的上下包络曲线,其插值算法会引起过冲、欠冲和不完全包络等问题。为此提出了采用分段三次多项式贝塞尔插值算法作为EMD分解过程中的包络算法,从而减小分解过程中的误差,准确提取非平稳信号。最后,利用Mat lab软件进行仿真实验,结果证明能够有效的改进EMD中的曲线包络中的问题和边界效应。

Empirical mode decomposition （EMD） is a signal analysis method which has received much attention lately due to its performance in a number of applications. But as a new signal processing methods, it is lacking a theoretical foundation and therefore, our understanding of it has come through intuition and experimental validation. The traditional EMD algorithm adopts the cubic spine interpolation as an effective tool to process nonstationary signal, but it can not effectively extract the characteristic frequencies from a highly non - stationary signal, and the over- shoots and the undershoots may become a common phenomenon during the decomposition process. In order to solve the problem, the paper presents the Piecewisc Cubic Bessel interpolation as a substitute for the spine interpolating. Finally, the simulation signal is used to verify the effectiveness of the improved EMD.