基于EEMD和伪信号技术的涡街信号处理方法  

Vortex flow signal processing method based on EEMD and masking signal technique

作　　者：张秀锋[1] 王勇[1] 

机构地区：[1]合肥工业大学机械工程学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2017年第11期1452-1457,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(41076061;51279044)

摘　　要：针对深海水平流涡街信号的微弱性、低频性,易被噪声淹没难以提取的特点,文章提出一种基于整体经验模态分解(ensemble empirical mode decomposition,EEMD)和伪信号技术的消噪方法。通过分析经过EEMD分解及希尔伯特变换(Hilbert transform,HT)后信号的时频特性,确定组成信号的主要两阶固有模态参数(intrinsic mode functions,IMF),结合双伪信号技术进一步处理后识别信号频率。模拟实验与仿真结果对比表明,该方法能有效提高信号的抗干扰能力,精确估计涡街信号频率。Vortex flow signal of deep horizontal flow is featured with weak intensity and low frequency and it is easy to be masked by noise .To solve these problems ,a new denoising method based on en-semble empirical mode decomposition（EEMD） and masking signal technique was proposed .Through time-frequency analysis of the noise signal after EEMD decomposition and Hilbert transform （HT ） , the major two intrinsic mode functions（IMF） were determined .Then the frequency of signal could be identified through further processing step using double masking signals technique .By comparing the simulation results with experimental results ,it is concluded that the proposed method effectively im-proves the anti-interference property of the signal and provides accurate estimate of vortex flow signal f requency .

关 键 词：希尔伯特-黄变换(HHT) 经验模态分解(EMD) 趋势项 微弱信号 涡街信号 

分 类 号：TP274.2[自动化与计算机技术—检测技术与自动化装置]