桥梁索塔GPS监测信号的小波分析  

Wavelet analysis of GPS monitoring signal of bridge cable tower

作　　者：戴建彪 岳东杰[1] 汤同旭 陈健[1] DAI Jianbiao;YUE Dongjie;TANG Tongxu;CHEN Jian(School of Earth Science and Engineering,Hohai University,Nanjing 211100,China)

机构地区：[1]河海大学地球科学与工程学院

出　　处：《测绘科学》2019年第9期151-157,共7页Science of Surveying and Mapping

摘　　要：针对EMD分解后直接舍去高频含噪信号可能会导致有用信息的丢失,以及小波阈值去噪中采用全局阈值去噪效果不完善等问题,该文提出了一种基于SSA的改进EMD-Wavelet耦合模型运用于桥梁索塔GPS监测信号去噪。即先对原始监测信号进行奇异谱分析,提取信号的趋势项和周期项,分析不同阶段的信号特性从而对其进行合理分段。在对经过EMD分解后的高频信号进行小波去噪时,根据信号的分段结果和给定的阈值计算函数进行分段分层取阈值。结果表明,该方法能很好地对原始坐标序列进行降噪,并且各项评价指标均优于SSA重构去噪法、EMD分解去噪法和EMD-Wavelet全局阈值去噪法,去噪效果更佳,这为索塔监测数据信号提取提供了有意义的参考。The direct rounding of high frequency signals after empirical mode decomposition(EMD)may lead to the loss of useful information,and the imperfect effect of the wavelet threshold de-noised with aglobal threshold,an improved EMD-wavelet coupling model based on singular spectrum analysis(SSA)was proposed to de-noised the GPS monitoring signal of bridge pylons in the paper.First,the SSA of the original monitoring signal was carried out,the trend term and the periodic term were extracted,and the signal characteristics at different stages were analyzed to reasonably segment the signal.When the high frequency signal decomposed by EMD was de-noised by wavelet,the segmentation threshold was segmented according to the piecewise result of the signal and the function of the given threshold.The results showed that this method could reduce the noise of the original coordinate sequence well,and each evaluation index was better than three pre-existing model namely SSA,EMD and EMD-wavelet,this provided a meaning reference.

关 键 词：桥梁索塔GPS监测 奇异谱分析 经典模态分解 小波变换 去噪 

分 类 号：P227[天文地球—大地测量学与测量工程]