基于VMD及能量熵的隧道爆破振动信号降噪方法  

Noise Reduction Method of Tunnel Blasting Vibration Signal Based on VMD and Energy Entropy

作　　者：叶海旺[1,2] 钟航 周汉红 欧阳枧 龙桂华 雷涛[1,2] 李立峰[1,2] 王炯辉 赵明生[6] 余红兵 YE Haiwang;ZHONG Hang;ZHOU Hanhong;OUYANG Jian;LONG Guihua;LEI Tao;LI Lifeng;WANG Jionghui;ZHAO Mingsheng;YU Hongbing(School of Resources and Environment Engineering,Wuhan University of Technology Wuhan 430070,China;Hubei Key Laboratory of Mineral Resources Processing and Environment,Wuhan 430070,China;Security Administration of Wuhan Public Security Bureau,Wuhan 430077,China;Shenzhen Municipal Engineering Corporation,Shenzhen 518000,China;Minmetals Exploration and Development Co.,Ltd.,Beijing 100044,China;Poly Xinlian Blasting Engineering Limited Company,Guiyang 550002,China)

机构地区：[1]武汉理工大学资源与环境工程学院,湖北武汉430070 [2]矿物资源加工与环境湖北省重点实验室,湖北武汉430070 [3]武汉市公安局治安管理局,湖北武汉430077 [4]深圳市市政工程总公司,广东深圳518000 [5]五矿勘查开发有限公司,北京100044 [6]保利新联爆破工程集团有限公司,贵州贵阳550002

出　　处：《金属矿山》2024年第11期111-117,共7页Metal Mine

基　　金：湖北省重点研发计划项目(编号:2021BCA152);“十三五”国家重点研发计划项目(编号:2020YFC1909602);“十四五”国家重点研发计划项目(编号:2021YFC2902901);国家自然科学基金(青年科学基金)项目(编号:52104098)。

摘　　要：为从监测信号中获得真实的隧道爆破振动信号,引入一种基于变分模态分解(Variational Mode Decomposition,VMD)及能量熵的隧道爆破振动信号降噪方法。首先,在不同K值条件下对监测信号进行变分模态分解,得到一系列本征模态分量(Intrinsic Mode Function,IMF),计算各模态分量与原信号的相关系数,根据相关系数确定最优K值;然后,对原信号进行变分模态分解,得到一系列中心频率由高到低的本征模态分量,计算每个模态分量的能量熵,确认噪声与信号的分界;最后将信号分量进行重构,得到降噪后的隧道爆破信号。将该信号降噪方法应用于实际爆破振动信号处理,结果表明:与经验模态分解(Empirical Mode Decomposition,EMD)降噪法相比,信噪比更高,剩余能量百分比更高,均方差更低,反映出所提降噪方法在隧道爆破振动信号处理中具有一定的适用性。In order to obtain the real tunnel blasting vibration signal from the monitoring signal,a denoising method for tunnel blasting vibration signal based on Variational Mode Decomposition(VMD)and energy entropy is introduced.Firstly,the monitoring signal is decomposed by variational modes under different K values to obtain a series of Intrinsic Mode Functions(IMF),and the correlation coefficient between each intrinsic mode function and the original signal is calculated,and the optimal K value is determined based on the correlation coefficients.Secondly,the original signal is decomposed by variational modes to obtain a series of intrinsic mode functions with center frequencies from high to low,and the energy entropy of each intrinsic mode function is calculated to confirm the boundary between noise and signal.Finally,the signal components are reconstructed and the denoised tunnel blasting signal is obtained.The results show that,compared with Empirical Mode Decomposition(EMD)noise reduction method,the signal-to-noise ratio is higher,the percentage of residual energy is also higher,and the mean square error is lower,which shows that the proposed noise reduction method has certain applicability in tunnel blasting vibration signal processing.

关 键 词：隧道爆破振动信号 变分模态分解 能量熵 降噪 

分 类 号：TD235[矿业工程—矿井建设]