摘要
地震资料采集获得的地震信号资料中往往夹杂着大量的噪音信息,地震信号的有效去噪对后续的地震资料解释具有重要的意义。本文提出一种基于经验模态分解(Empirical Mode Decomposition,EMD)和改进小波阈值的地震信号去噪方法。将含噪地震信号分解为一系列本征模态函数(Intrinsic Mode Function,IMF),并根据自相关理论,对从IMFs中筛选的高频分量应用改进的小波高参数阈值算法处理,低频IMF分量应用改进的小波低参数阈值处理,最后对处理后的IMFs进行重构。利用本文提出的地震信号去噪方法对合成地震信号和实际地震信号进行去噪处理,并与EMD去噪效果进行对比,结果均表明本文所提方法的去噪效果优于常规的EMD方法。
The seismic signal data obtained during the seismic data acquisition process often contain a lot of interference information, which seriously affects the subsequent interpretation work. The quality of the seismic signal denoising directly affects the application effect. In this paper, a seismic denoising algo-rithm based on Empirical Mode Decomposition(EMD) and improved wavelet threshold is proposed. The EMD algorithm is applied to the noisy signal to decompose it into a series of Intrinsic Mode Functions(IMF). According to the autocorrelation theory, the high-frequency components selected from the IMFs are processed by the improved wavelet high-parameter threshold algorithm. Apply improved wavelet low parameter threshold processing, and finally reconstruct the processed IMFs. Denoising the synthetic and actual seismic records separately, and comparing it with the EMD denoising algorithm, the results show that this method is superior to the conventional EMD denoising algorithm.
作者
巨鑫
郑小鹏
武科含
周健
商冬明
徐静霞
Xin Ju;Zhen xiaopeng;Kehan Wu;Jian Zhou;Dongming Shang;Jingxia Xu(China University of Mining and Technology,School of Resources and Geosciences,Xuzhou 221116;Sinopec Geophysical Research Institute,Nanjing 210013)
出处
《内蒙古石油化工》
CAS
2020年第5期44-49,共6页
Inner Mongolia Petrochemical Industry
基金
2019年中国矿业大学大学生创新训练计划国家级项目“改进EMD的地震信号去噪方法研究及应用”(编号:201910290035Z)
国家自然科学基金企业创新发展联合基金“海相深层油气富集机理与关键工程技术基础研究”(编号:U19B6003)联合资助
关键词
地震信号
经验模态分解
小波阈值去噪
去噪
信噪比
Seismic signals
Empirical mode decomposition
Wavelet threshold
denoising
Signal-to-noise ratio
作者简介
巨鑫(1999-),男,中国矿业大学资源学院地球物理学专业学生。
