经验模式分解回顾与展望

发布时间：2018-03-15 01:33

本文选题：经验模式分解　切入点：固有模态函数　出处：《计算机工程与科学》2014年01期 　论文类型：期刊论文

【摘要】：经验模式分解EMD打破了Fourier变换、小波分解等传统数据分析方法需要预先设定基函数的局限,是一种完全由数据驱动的自适应非线性非平稳时变信号分解方法,可以将数据从高频到低频分解成具有物理意义的少数几个固有模态函数分量和一个余量。首先介绍了原始EMD方法的原理和算法;接着,总结归纳了EMD当前的研究现状,分析了EMD存在的端点效应、模态混叠、运行速度问题及其在二维情况下的问题并对国内外学者解决这些问题的方法进行了概述和比较;最后结合EMD研究存在的难题指出了EMD进一步研究与应用的发展方向。
[Abstract]:Empirical mode decomposition (EMD) breaks the limitation of traditional data analysis methods such as Fourier transform wavelet decomposition and so on which need to set the basis function in advance. It is a data-driven adaptive nonlinear non-stationary time-varying signal decomposition method. The data can be decomposed from high frequency to low frequency into a few intrinsic modal function components and a residue with physical significance. Firstly, the principle and algorithm of the original EMD method are introduced, and then the current research status of EMD is summarized. In this paper, the endpoints effect, modal aliasing, running speed and their problems in two dimensions of EMD are analyzed, and the methods used by scholars at home and abroad to solve these problems are summarized and compared. Finally, the development direction of further research and application of EMD is pointed out in combination with the problems existing in the study of EMD.
【作者单位】：陕西师范大学计算机科学学院;
【基金】：国家自然科学基金资助项目(51275380) 陕西省科学技术研究发展计划项目(2012K06-36) 陕西师范大学中央高校基本科研业务费(GK201102006)
【分类号】：TN911.7

【参考文献】