基于奇异值分解的角度域地震波束形成方法

2017年 56卷 第No. 5期
阅读:95
查看详情
Angle domain seismic beam-forming method based on singular value decomposition
(1.地球信息探测仪器教育部重点实验室(吉林大学),吉林长春130026;2.吉林大学仪器科学与电气工程学院,吉林长春130026;3.吉林大学地球科学学院,吉林长春130061)
(1.Key Laboratory of Geophysical Exploration Equipment,Ministry of Education,Jilin University,Changchun 130026,China; 2.College of Instrumentation & Electrical Engineering,Jilin University,Changchun 130026,China;3.College of Earth Sciences,Jilin University,Changchun 130061,China)

基于接收阵列的时域地震波束形成方法在处理低信噪比地震记录时,因叠加记录数量过大,导致有效波束宽度不能覆盖全部接收排列,波束宽度外的信号产生畸变。而且,当地震记录中相干噪声和随机噪声很强时,该方法压制噪声不均匀,降噪效果不佳。为此,针对相干噪声和随机噪声强、目标信号弱的地震记录,提出了一种基于奇异值分解的角度域地震波束形成方法。该方法根据地震记录中噪声与信号之间的相干性差异,利用角度域地震波束形成方法增强整个反射信号的能量和相干性,利用奇异值分解方法压制相干噪声和随机噪声,从降噪和增强有效信号的双重处理上,较大程度地提高地震记录的信噪比。仿真实验和实测信号分析结果表明,该方法不仅能增强有效信号,还能进一步压制记录中相干噪声和随机噪声,明显提高了地震记录的信噪比。

When a time-domain seismic beam-forming method based on the receiving array processes low signal-to-noise ratio (SNR) seismic records,owing to the large number of stack records,the effective beam width can not cover all the receiving arrangements,resulting in signal distortion outside the beam width.In addition,when encountering seismic records with strong coherent noise and random noise,the above method suppresses the noise unevenly,resulting in poor denoising effect.In this paper,for seismic records with strong coherent noise,random noise,and weak target signals,an angle domain seismic beam-forming method based on singular value decomposition is proposed.According to the correlation difference between the noise and the signal in seismic records,this method enhances the energy and correlation of the whole reflection signal by using an angle domain seismic beam-forming method,and suppresses the coherent noise and random noise by singular value decomposition (SVD).By considering both the noise reduction and the signal enhancement,the method can effectively improve the signal-to-noise ratio of the seismic records.The tests results of simulation experiment and the measured signal showed the effectiveness of the proposed method.

角度域划分; 波束形成; 奇异值分解; 去噪; 信噪比;
angle domain division,; beam-forming,; singular value decomposition,; denoising,; SNR;

深部探测技术与实验研究专项SinoProbe09-06(201311197,201011083)、高等学校博士学科专项科研基金(20110061110053)联合资助。

10.3969/j.issn.1000-1441.2017.05.007