去除爆炸冲击波信号高频噪声的联合处理方法  

A Joint Processing Method of Removing High Frequency Noise from Shock Wave Signal

作　　者：刘宇[1,2] 李新娥 崔春生[1,2] 李顺 LIU Yu;LI Xine;CUI Chunsheng;LI Shun(State Key Laboratory of Dynamic Measurement Technology,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]中北大学省部共建动态测试技术国家重点实验室,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：《探测与控制学报》2023年第2期61-66,共6页Journal of Detection & Control

摘　　要：爆炸场采集的冲击波信号存在高频噪声干扰,为了提高滤除高频噪声干扰的精度,提出了希尔伯特黄变换(HHT)与Bessel联合的信号处理方法。采用经验模态分解(EMD)得到一系列固有模态函数(IMF)和余量,通过对各IMF与原始信号互相关系数的分析,初步判断出含噪IMF,并利用希尔伯特变换(HT)确定含噪IMF;同时,采用Bessel低通数字滤波器对含噪IMF进行处理;最后,将Bessel处理后的IMF、纯净的IMF和余量进行重构得到处理后的信号。试验结果表明:与目前已有的爆炸冲击波信号处理方法对比,该方法信噪比更高,均方根误差更小,超压峰值和原始信号的超压峰值较为贴近,去噪精度更高,去除了高频干扰,保留了0~70 kHz反映冲击波本征特点的信号,提高了信号分析的准确性。The shock wave signal collected in the explosion field has high-frequency noise interference.To improve the accuracy of filtering high-frequency noise interference,a joint signal processing method of HHT(Hilbert-Huang transform)and Bessel was proposed.A series of IMF(intrinsic mode function)components and residual were obtained by EMD(empirical mode decomposition).Through the analysis of the correlation between each IMF and the original signal,the noise IMF components were preliminarily judged,and the noise IMF components are determined by HT(Hilbert transform).At the same time,Bessel low-pass digital filter was used to process noise IMF components.Finally,the IMF components processed by Bessel,pure IMF components and residual were reconstructed to obtain the processed signal.The results showed that compared with the existing explosion shock wave signal processing methods,this method had higher SNR(signal noise ratio),smaller RMSE(root-mean-square error),closer overpressure to the overpressure of the original signal,higher denoising accuracy,removed high-frequency interference and retained the signal reflecting the intrinsic characteristics of shock wave at 0~70 kHz.

关 键 词：信号处理 希尔伯特黄变换 Bessel滤波器 经验模态分解 小波阈值 

分 类 号：O384[理学—流体力学]