基于CEEMDAN和改进小波模极大值的高低频混合信号去噪算法  

High and Low Frequency Mixed Signal Denoising Algorithm Based on CEEMDAN and Improved Wavelet Modulus Maximum

作　　者：王轶楷 谢锐[1,2] WANG Yikai;XIE Rui(Science and Technology on Electronic Test and Measurement Laboratory,North University of China,Taiyuan 030051,China;School of Electrical and Control Engineering,North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学电子测试技术重点实验室,山西太原030051 [2]中北大学电气与控制工程学院,山西太原030051

出　　处：《中北大学学报（自然科学版）》2023年第4期366-373,共8页Journal of North University of China(Natural Science Edition)

基　　金：山西省自然科学基金资助项目(20210302123048)。

摘　　要：利用容栅传感器获取的传动轴低频扭矩信号远距离传输时需将其加载到高频信号上,而系统在复杂的应用环境下所检测出的这种高低频混合信号混杂了环境噪声和系统电磁噪声,该混合信号信噪比低,提取困难,对此本文提出了一种基于CEEMDAN-改进小波模极大值去噪算法。该算法先利用CEEMDAN进行分解,然后对CEEMDAN分解后的固有模态分量(IMF)进行改进小波模极大值去噪,进而逼近有用信号。利用CEEMDAN去噪、小波模极大值去噪和本文算法来对高低频混合模拟信号进行去噪,并应用信噪比和均方根误差指标来评价去噪效果,仿真结果表明,与CEEMDAN去噪和小波模极大值去噪相比,本文算法的信噪比分别提升了46.9%和35.9%,均方根误差分别减小了98.4%和97.2%。利用CEEMDAN-小波阈值去噪算法、EEMD-小波模极大值去噪算法和本文算法对实测信号进行去噪,实验结果表明本文算法的去噪效果更好,去噪后的信号更光滑,更具有稳定性。The low frequency torque signal of the drive shaft detected by the capacitive grating sensor was loaded onto the high frequency signal to achieve long-distance transmission.In a complex application environment,the high and low frequency signals detected by the system were mixed with environmental noise and system electromagnetic noise.Aiming at the signal questions of extracting difficulty and low signal-to-noise ratio,a denoising algorithm based on CEEMDAN-improved wavelet modulus maxima was proposed.The algorithm first used CEEMDAN for decomposition,and then the intrinsic mode function(IMF)through CEEMDAN decomposition was denoised by using improved wavelet modulus maximum denoising to approximate the useful signal.The high and low frequency mixed analog signals were denoised by CEEMDAN denoising,wavelet modulus maxima denoising and the proposed algorithm,and the signal-to-noise ratio and root mean square error index were used to evaluate the denoising effect of three algorithms.Compared with CEEMDAN denoising and wavelet modulus maxima denoising,the signal-to-noise ratio of the proposed algorithm is improved by 46.9%and 35.9%respectively,and the root mean square error is reduced by 98.4%and 97.2%respectively.The real measured signals were denoised by CEEMDAN-wavelet threshold denoising algorithm,EEMD-wavelet modulus maxima denoising algorithm and the proposed algorithm.The experimental results show that the denoising effect of the proposed algorithm is better,and the denoised signal is smoother and more stable than other algorithms.

关 键 词：CEEMDAN 小波模极大值 高低频混合信号 去噪 

分 类 号：TP333[自动化与计算机技术—计算机系统结构]