基于CEEMD-小波阈值的低频振动信号降噪研究  

Research on Noise Reduction of Low-Frequency Vibration Signal Based on CEEMD-Wavelet Threshold

作　　者：孙旭 熊鸣[1] 王丽婕[1] SUN Xu;XIONG Ming;WANG Lijie(College of Automation,Beijing Information Science and Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学自动化学院,北京100192

出　　处：《信息与电脑》2024年第23期49-53,共5页Information & Computer

摘　　要：针对低频振动信号在采集过程中混入噪声的问题,文章提出了一种基于互补集合经验模态分解(Complementary Ensemble Empirical Mode Decomposition,CEEMD)和小波阈值的降噪方法,来对低频振动信号进行降噪处理。先将光纤光栅加速度传感器采集的低频振动信号进行CEEMD分解得到高频固有模态函数(Intrinsic Mode Function,IMF)分量,再使用小波阈值算法对分解后的IMF分量进行降噪,最后把降噪后的高频IMF分量与低频IMF分量信号进行重构,得到需要的降噪信号。将CEEMD-小波阈值降噪算法与传统的EEMD和CEEMD算法进行仿真对比可知,CEEMD-小波阈值降噪算法拥有较好的降噪效果。Aiming at the problem of low-frequency vibration signals mixed with noise in the acquisition process,the article proposes a noise reduction method based on complementary ensemble empirical mode decomposition(CEEMD)and wavelet thresholding to reduce the noise of low-frequency vibration signals.The low-frequency vibration signal collected by the fiber grating acceleration sensor is first decomposed by CEEMD to obtain the high-frequency intrinsic mode function(IMF)component,and then the wavelet thresholding algorithm is used to perform noise reduction on the decomposed IMF component,and then the noise-reduced high-frequency IMF component and low-frequency IMF component signals are reconstructed to obtain the desired noise-reduced signal.Comparing the CEEMD-wavelet thresholding algorithm with the traditional EEMD and CEEMD algorithms,it can be seen that the CEEMD-wavelet thresholding algorithm has a better noise reduction effect.

关 键 词：CEEMD-小波阈值降噪 低频振动 光纤光栅 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]