基于最优特征的改进经验模态分解方法  

Improved empirical mode decomposition method based on optimal feature

作　　者：刘薇娜[1] 周小龙[1] 姜振海[2] 马风雷[2] 

机构地区：[1]长春理工大学机电工程学院,长春130022 [2]长春工业大学机电工程学院,长春130012

出　　处：《吉林大学学报（工学版）》2017年第6期1957-1963,共7页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(51075041);吉林省教育厅"十二五"科学技术研究重点项目(J00-JYT2014-12)

摘　　要：针对经验模态分解(EMD)过程中存在的模态混叠等问题,提出了一种基于最优特征的自适应白噪声平均总体经验模态分解方法。该方法采用基于边界局部均值延拓的方法抑制端点效应问题,同时,在经验模态分解的每个阶段自适应地添加有限次白噪声,保证在平均次数相对少的情况下,通过计算唯一残余分量来获取信号的固有模态函数,从而避免了模态混叠问题的产生。通过分析仿真信号和实测信号,证明了该方法对模态混叠现象有一定的抑制作用,同时可有效避免端点效应问题的产生。To solve the problem that mode mixing exists during Empirical Mode Decomposition （EMD）, an improved EMD method based on the best feature of intrinsic mode function is proposed. First, the method of boundary local mean is used to deal with the end effect problem. Meanwhile, limited white noises are added at each stage of the EMD. Then, a unique residual component is computed to get each intrinsic mode function with a small number of ensemble size, thus, the mode mixing problem is solved. The results of the numerical simulation and real experiments show that the proposed method has good performance in solving the mode mixing problem and can restrain the end effect problem effectively.

关 键 词：机械设计 经验模态分解 最优特征 模态混叠 固有模态函数 

分 类 号：TN911.7[电子电信—通信与信息系统]