城市地铁区间隧道爆破振动信号趋势项和噪声消除方法  

Trend term removal and de-noising method for blasting vibration signal of urban subway tunnel excavation

作　　者：付晓强 刘纪峰[1,2] 黄凌君 蔡雪霁 王军芳 刘静[1,2] FU Xiaoqiang;LIU Jifeng;HUANG Lingjun;CAI Xueji;WANG Junfang;LIU Jing(School of Civil Engineering,Sanming University,Sanming 365004,China;Key Laboratory of Engineering Material&Structure Reinforement in Fujian Province College(Sanming University),Sanming 365004,China)

机构地区：[1]三明学院,建筑工程学院,福建三明365004 [2]工程材料与结构加固福建省高等学校重点实验室,福建三明365004

出　　处：《铁道科学与工程学报》2020年第9期2328-2337,共10页Journal of Railway Science and Engineering

基　　金：福建省中青年教师教育科研项目(JAT190697);三明市引导性科技项目计划(2019-S-28);三明学院引进高层次人才科研启动经费资助(18YG13);隧道软弱围岩技术研究(2-JF-2019-浦炎高速YA18-2-002)。

摘　　要：为消除隧道爆破振动信号中易存在的趋势项和噪声干扰,提出了局部均值分解与基线估计和稀疏化去噪方法。选取含有基线漂零的爆破信号并采用LMD将其分解为若干独立分量,对各分量进行BEADS运算得到剔除趋势项和随机高频噪声的校正信号。将各校正信号进行重构获得消除趋势项的特征信号,对其引入噪声特征进行人工判别并二次消除其中的重构引入高频噪声得到真实信号。分析结果表明:爆破信号中含有的趋势项往往位于信号低频段,而随机噪声和引入噪声干扰通常位于高频段,组合算法可根据信号特点自适应地消除二者对信号特征提取精度的影响,从而准确揭示信号本质特征属性。实践证明该算法收敛速度快,是一种高效便捷的信号预处理方法。In order to eliminate the trend term and noise problems in the vibration signal of tunnel blasting,a method of local mean decomposition,baseline estimation and sparse de-noising was proposed.The blasting signal containing baseline drift was selected and decomposed into several independent components by LMD.The BEADS were carried out in each component to obtain the de-trending signals that eliminate the trend term and high-frequency random noise.Each corrected component was reconstructed to obtain the characteristic signal that eliminates the trend term,and the introducing noise feature of characteristic signal was manually distinguished and the introducing high frequency noise was removed second time to obtain the real signal.The results show that the trend term contained in the blasting signal is usually located in the low frequency band,while the random and introducing noise interference is usually located in the high frequency ones.It is proved that the algorithm has fast convergence speed and is an efficient and convenient pre-processing method of signal.

关 键 词：地铁隧道 爆破振动 信号趋势项 高频噪声 相关性分析 

分 类 号：TD235.1[矿业工程—矿井建设]