方向导数迹变换面波压制  被引量:5

Ground-roll attenuation using the directional derivative trace transform

在线阅读下载全文

作  者:聂鹏飞[1] 李月[1] 曾谦[1] 林洪波[1] 吴宁[1] 

机构地区:[1]吉林大学通信工程学院,长春130026

出  处:《地球物理学报》2012年第6期2035-2043,共9页Chinese Journal of Geophysics

基  金:国家自然科学基金(40974064);国家自然科学基金重点项目(41130421)资助

摘  要:面波是地震勘探中常见的一种相干干扰,它的存在严重的影响着地震记录的信噪比.由于面波和有效波具有相关性且面波的频带和有效波的频带总有重叠的部分,在时域或频域二者不能明显分开,因此在时域或频域采用切除法压制面波会造成子波畸变和有效信息的损失.本文提出一种利用方向导数迹变换压制面波的新方法.文中推导了方向导数迹变换的反变换公式.地震记录的方向导数迹变换(Directional Derivative Trace Transform,DDTT)由两部分组成,一部分主要体现面波,能量集中;另一部分主要体现反射有效波,能量相对分散.根据这两部分能确定压制面波的阈值,通过这一阈值在正变换中压制面波后,再通过反变换返回时-空域就可达到压制面波的目的.理论和实际数据的处理都取得了令人满意的效果,表明了本文提出方法的可行性和有效性.Ground-roll is common and mainly coherent noise in seismic surveys for a significant reduction in the signal-to-noise ratio. Since the ground-roll is coherent with reflections and the spectrums of them are always overlapping. Surgical muting as a common way to filtering ground- roll will cause waveform distortions and information missing. We present a novel method, namely directional derivative trace transform to attenuate ground-roll. The forward and inverse transform are given and derived in this paper. The directional derivative trace transform of seismic records contains two entirely different parts: one part is mainly representing ground-roll where the energy is highly centralized, the other shows reflections where energy is obviously scattered. According to these two parts, we can determine a threshold which can suppress the ground-roll in the domain of directional derivative trace transform. Then the purpose of ground-roll attenuation can be achieved by the inverse directional derivative trace transform to make reflections back to time-space domain. The feasibility and effectiveness of the proposed method is confirmed by implementation on simulations and field data.

关 键 词:方向导数迹变换与反变换 面波 信噪比 地震勘探 

分 类 号:P631[天文地球—地质矿产勘探]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象