[1]余华龙,徐国林.基于MEEMD和时变VARMA模型的地震动噪声清除及基线修正[J].地震工程与工程振动,2022,42(02):172-180.[doi:10.13197/j.eeed.2022.0217]
 YU Hualong,XU Guolin.Earthquake signal noise filtering and baseline correction based on MEEMD and time-varying VARMA model[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2022,42(02):172-180.[doi:10.13197/j.eeed.2022.0217]
点击复制

基于MEEMD和时变VARMA模型的地震动噪声清除及基线修正
分享到:

《地震工程与工程振动》[ISSN:/CN:]

卷:
42
期数:
2022年02期
页码:
172-180
栏目:
论文
出版日期:
2022-04-30

文章信息/Info

Title:
Earthquake signal noise filtering and baseline correction based on MEEMD and time-varying VARMA model
作者:
余华龙 徐国林
西南林业大学 土木工程学院, 云南 昆明 650224
Author(s):
YU Hualong XU Guolin
School of Civil Engineering, Southwest Forestry University, Kunming 650224, China
关键词:
改进的集合经验模态分解时变VARMA模型基线修正
Keywords:
MEEMDtime-varying VARMA modelsbaseline correction
分类号:
P315.9
DOI:
10.13197/j.eeed.2022.0217
摘要:
受背景噪声、仪器倾斜以及外部环境因素产生的低频信号和直流分量影响,由原始地震动信号得到的位移信号会产生波形偏移(基线漂移)现象。目前,常用的地震动去噪方法有滤波法、多项式拟合法、经验模态分解(简称EMD)去噪法、小波去噪法等,但上述方法均有局限性。在探索地震动信号基线修正和去噪过程中,以改进的集合经验模态分解(简称MEEMD)为基础结合时变向量自回归模型(简称VARMA)模型,对信号进行预处理和基线修正,利用排列熵算法除去随机性较大噪声信号,将剩余信号通过EMD分解得到的本征模态函数(IMF)利用时变VARMA模型还原地震动信号,开展函数积分运算获得保留永久位移的位移信号。对比文中方法与小波分解截断基线修正法处理同一地震动信号得到的位移信号发现,文中方法能够较好的处理目标信号里面包含的高频干扰噪声信号和低频噪声信号,获得保留永久位移的位移信号,精度可满足工程实践的需要。
Abstract:
Affected by the background noise,instrument tilt,and low-frequency signals and DC components generated by external environmental factors,the displacement signal obtained from the original ground motion signal will cause a waveform shift(baseline drift)phenomenon. At present,the commonly used ground motion denoising methods include filtering method,polynomial fitting method,empirical mode decomposition(EMD)denoising method, wavelet denoising method,etc.,but the above methods have their limitations. In the process of exploring the baseline correction and denoising of ground motion signals,the author combines the time-varying vector autoregressive model(VARMA)model based on the improved ensemble empirical mode decomposition(MEEMD)to preprocess and correct the baseline. The permutation entropy algorithm is used to remove the large random noise signal,the eigenmode function(IMF)obtained by the EMD decomposition of the remaining signal is used to restore the ground motion signal using the time-varying VARMA model,and the function integration operation is carried out to obtain the displacement signal which retain the permanent displacement. Through the method and wavelet decomposition truncated baseline correction method to process the displacement signal obtained from the same ground motion signal,it is found that the method in this paper can better process the high-frequency interference noise signal and low-frequency noise signal contained in the target signal,and obtain the displacement signal that retains the permanent displacement. The accuracy can meet the needs of engineering practice.

参考文献/References:

[1] 刘继承,徐庆华,查建新. 用加速度传感器测量振动位移的方法[J]. 现代雷达,2007,29(5):69-71. LIU Jicheng,XU Qinghua,ZHA Jianxin. The method of measuring vibration displacement with acceleration sensor[J]. Modern Radar,2007,29(5):69-71.(in Chinese)
[2] 周英杰. 加速度测试积分位移算法及其应用研究[D]. 重庆:重庆大学,2013. ZHOU Yingjie. Acceleration Test Integral Displacement Algorithm and Its Application Research[D]. Chongqing:Chongqing University,2013. (in Chinese)
[3] 顾名坤,吕振华. 基于振动加速度测量的振动速度和位移信号识别方法探讨[J]. 机械科学与技术,2011,30(4):522-526. GU Mingkun,LV Zhenhua. Discussing the identification method of vibration velocity and displacement signal based on vibration acceleration measurement[J]. Miechanical Science and Technology for Aerospace Engineering,2011,30(4):522-526.(in Chinese)
[4] 肖梅玲,高春涛,叶燎原. 不同小波基下的结构位移反应动力可靠性分析[J]. 世界地震工程,2002,18(2):163-166. XIAO Meiling,GAO Chuntao,YE Liaoyuan. Dynamic reliability analysis of structural displacement response under different wavelet bases[J]. World Earthquake Engineering,2002,18(2):163-166.(in Chinese)
[5] 林金朝,刘乐乐,李国权,等. 基于改进EMD的心电信号基线漂移消除方法[J]. 数据采集与处理,2018,33(5):880-890. LIN Jinchao,LIU Lele,LI Guoquan,et al. ECG baseline drift elimination method based on improved EMD[J]. Data Acquisition and Processing, 2018,33(5):880-890.(in Chinese)
[6] 哈克. 经验模态分解法简析[EB/OL].(2013-03-30)http://blog.sina.com.cn/s/blog_55954cfb0102e9y2.html. HACK. Brief analysis of empirical mode decomposition method[EB/OL].(2013-03-30)http://blog.sina.com.cn/s/blog_55954cfb0102e9y2.html. (in Chinese)
[7] 程军圣,于德介,杨宇. 基于EMD的信号瞬时特征的小波分析方法[J]. 地震工程与工程振动,2004,24(2):181-186. CHENG Junsheng,YU Dejie,YANG Yu. Wavelet analysis method of signal instantaneous characteristics based on EMD[J]. Earthquake Engineering and Engineering Dynamics,2004,24(2):181-186.(in Chinese)
[8] 董银峰,李英民,赖明,等. 基于EMD和VARMA模型的地震动仿真[J]. 地震工程与工程振动,2007,23(7):15-21. DONG Yinfeng,LI Yingmin,LAI Ming,et al. Ground motion simulation based on EMD and VARMA models[J]. Earthquake Engineering and Engineering Dynamics,2007,23(7):15-21.(in Chinese)
[9] 李英民,董银峰,夏洪流. 考虑频率非平稳特性的地震波ARMA模型仿真方法[J]. 重庆建筑大学学报,2000,22(增刊1):123-127. LI Yingmin,DONG Yinfeng,XIA Hongliu. A seismic wave ARMA model simulation method considering frequency non-stationary characteristics[J]. Journal of Chongqing Jianzhu University,2000,22(S1):123-127.(in Chinese)
[10] 董银峰. 非平稳信号处理方法的改进及在地震工程中的应用研究[D]. 重庆:重庆大学,2010. DONG Yinfeng. Improvement of Non-stationary Signal Processing Method and Its Application in Earthquake Engineering[D]. Chongqing:Chongqing University,2010.(in Chinese)
[11] 张尚荣,毛杰明,惠迎新,等. 基于小波变换的层间隔震结构地震响应分析[J]. 地震工程学报,2017. 39(5):836-842. ZHANG Shangrong,MAO Jieming,HUI Yingxin,et al. Seismic response analysis of story-isolated structure based on wavelet transform[J]. Chinese Journal of Earthquake Engineering,2017,39(5):836-842.(in Chinese)
[12] 叶燎原,肖梅绮. 基于小波变换的地震反应分析[J]. 世界地震工程,2001,17(2):14-17. YE Liaoyuan,XIAO Meiqi. Seismic response analysis based on wavelet transform[J]. World Earthquake Engineering,2001,17(2):14-17.(in Chinese)
[13] 许鑫,史治宇,龙双丽,等. 利用加速度响应连续小波变换的时变系统物理参数识别[J]. 振动工程学报,2013,26(1):8-14. XU Xin,SHIZhiyu,LONG Shuangli,et al. Physical parameter identification of time-varying systems using acceleration response continuous wavelet transform[J]. Journal of Vibration Engineering,2013,26(1):8-14.(in Chinese)
[14] 李锦飞,李人厚,刘贵忠. 基于小波多分辨分析的极化分析和滤波方法[J]. 信号处理,1999,15(1):88-92. LI Jinfei,LI Renhou,LIU Guizhong. Polarization analysis and filtering method based on wavelet multi-resolution analysis[J]. Signal Processing, 1999,15(1):88-92.(in Chinese)
[15] 张慧娟,李冬. 相关系数判决的EMD在振动数据趋势项提取中的应用[J]. 舰船电子工程,2018(5):159-163. ZHANG Huijuan,LI Dong. Application of EMD with correlation coefficient judgment in vibration data trend item extraction[J]. Ship Electronic Engineering,2018(5):159-163.(in Chinese)
[16] 朱学锋,韩宁. 基于经验模态分解的非平稳信号趋势项消除[J]. 飞行器测控学报,2012(1):65-70. ZHU Xuefeng,HAN Ning. Non-stationary signal trend elimination based on empirical mode decomposition[J]. Journal of Aircraft Measurement and Control,2012(1):65-70.(in Chinese)
[17] 甘雨,隋立芬,戚国宾,等. 利用EMD阈值消噪改进MW组合周跳探测性能[J]. 大地测量与地球动力学,2015(4):666-670. GAN Yu,SUI Lifen,QI Guobin,et al. Using EMD threshold denoising to improve MW combined cycle slip detection performance[J]. Geodesy and Geodynamics,2015(4):666-670.(in Chinese)
[18] 东方. 基于离散小波变换的方波消噪[J]. 科技风,2015(11):101. DONG fang. Square wave denoising based on discrete wavelet transform[J]. Solicitation of Manuscripts,2015(11):101.(in Chinese)
[19] 王彩霞,刘义艳. 基于经验小波变换的结构损伤特征提取[J]. 计算机与数字工程,2020(1):189-193. WANG Caixia,LIU Yiyan. Structural damage feature extraction based on empirical wavelet transform[J]. Computer and Digital Engineering,2020(1):189-193.(in Chinese)
[20] 叶燎原,肖梅琦. 连续小渡变换用于结构地震反应分析[J]. 世界地震工程,2001,17(4):79-83. YE Liaoyuan,XIAO Meiqi. Continuous small cross transformation for structural seismic response analysis[J]. World Earthquake Engineering, 2001,17(4):79-83.(in Chinese)

备注/Memo

备注/Memo:
收稿日期:2021-2-17;改回日期:2021-8-30。
基金项目:国家自然科学基金项目(51768065)
作者简介:余华龙(1996-),男,硕士研究生,主要从事结构地震反应分析方法研究.E-mail:2438015460@qq.com
通讯作者:徐国林(1977-),男,教授,博士,主要从事土和结构相互作用和结构非线性地震反应分析研究.E-mail:346796145@qq.com
更新日期/Last Update: 1900-01-01