基于CEEMDAN与SVD的泄流结构振动信号降噪方法.docVIP

基于CEEMDAN与SVD的泄流结构振动信号降噪方法 张建伟 侯鸽 暴振磊 张翌娜 马颖 华北水利水电大学水利学院 黄河水利职业技术学院土木与交通工程学院 X 关注成功！ 加关注后您将方便地在 我的关注中得到本文献的被引频次变化的通知！ 新浪微博 腾讯微博 人人网 开心网 豆瓣网 网易微博 摘????要： 针对泄流结构振动信号非平稳性和特征信息被强噪声淹没的实际问题, 提出一种基于具有自适应噪声的完整集成经验模态分解 (CEEMDAN) 和奇异值分解 (SVD) 联合的信号降噪方法。对一维泄流振动信号时程进行CEEMDAN分解, 将信号分解为一系列固有模态函数分量 (IMF) , 运用频谱分析方法筛选包含主要振动信息的IMF分量, 滤除低频水流噪声, 实现信号的初次滤波;利用排列熵理论确定含噪声较多的IMF分量, 采用奇异值分解技术提取奇异值信息, 运用奇异熵增量定阶理论滤除IMF分量中的高频噪声, 实现信号的二次滤波;将包含结构振动信息的IMF分量重构, 得到泄流结构的工作特征信息。结合拉西瓦模型振动实测数据, 运用该方法进行计算分析, 滤除高频和低频噪声, 提取结构振动特征信息;结果表明该方法在泄流结构特征信息提取方面具有优越性, 可为泄流结构在线监测和安全运行提供依据。 关键词： 泄流结构; 集成经验模态分解; 排列熵; 奇异值分解; 特性信息; 作者简介：张建伟, 男, 博士, 副教授, 1979年生 收稿日期：2016-06-17 基金：国家自然科学基金 51408223) A signal de-noising method for vibration signals from flood discharge structures based on CEEMDAN and SVD ZHANG Jianwei HOU Ge BAO Zhenlei ZHANG Yina MA Ying School of Water Resources, North China University of Water Resources and Electric Power; School of Civil and Transportation Engineering, Yellow River Conservancy Technical Institute; Abstract： In order to solve the problems that flood non-stationary vibration signals from discharge structures are contaminated by strong noise, a signal de-noising method based on complete ensemble empirical mode decomposition with adaptive noise ( CEEMDAN) and singular value decomposition ( SVD) was proposed. Firstly, the signal was decomposed with CEEMDAN to obtain a series of intrinsic mode functions ( IMFs) . The spectrum analysis method was used to screen the IMF containing real physical meaning to filter out low frequency noise. Then the permutation entropy theory was used to determine the IMF containing more noise components, and the IMF was used to get appropriate feature matrix with the phase space reconstruction method. A part of white noises were filtered out by SVD and the singular entropy increment theory. Finally the IMF was reconstructed with characteristic information to achieve the de-noised signal through spectrum analysis. The Laxiwa arch dam wa