基于KL散度的EMD虚假分量识别方法研究-中国电机工程学报.PDF

第32 卷 第11 期 中 国 电 机 工 程 学 报 Vol.32 No.11 Apr.15, 2012 112 2012 年4 月15 日 Proceedings of the CSEE ©2012 Chin.Soc.for Elec.Eng. 0258-8013 (2012) 11-0112-06 TK 267 A 470 20 文章编号： 中图分类号： 文献标志码： 学科分类号： ⋅ 基于K-L 散度的EMD 虚假分量识别方法研究 珣 韩中合，朱霄 ，李文华 (华北电力大学能源动力与机械工程学院，河北省 保定市 071003) A False Component Identification Method of EMD Based on Kullback-leibler Divergence HAN Zhonghe, ZHU Xiaoxun, LI Wenhua (School of Energy and Power Engineering and Mechanical Engineering, North China Electric Power University, Baoding 071003, Hebei Province, China) ABSTRACT: Empirical mode decomposition (EMD) has been 人提出了一种新的信号处理方法——经验模态分 widely used in many nonlinear fields for its adaptive filtering 解(empirical mode decomposition，EMD)[1-2] 。EMD properties and adaptive multiresolution, but there will be false 方法把复杂数据信号自适应分解成瞬时频率有意 components in the decomposition process. To solve this 义的、有限个可以是幅度或频率调制的高频和低频 problem, a method of the false components identification of EMD based on Kullback-Leibler divergence was proposed. 的本征模态函数。EMD 方法简单、直观，具有很 First, the original signal was decomposed into several intrinsic 多优点，如：自适应性、完备性、近似正交性、本 mode functions (imf); then the Kullback-Leibler divergence 征模态函数imf 的调制性[3-5] 。EMD 方法是一种很 value between the original signal and each imf was calculated; 有力的非线性、非平稳信号分析工具，目前被广泛 finally the values were put in order. The imf, which 应用在机械故障诊断、地震、生物医学、语音增强、 corresponds to the big value, was regarded as a false 雷达等很多领域，取得了良好