摘要
在地震勘探中,随着炮道密度的快速提升和大块三维的不断部署,海量的炮工作量对井炮“交替式”激发效率提出了严峻挑战,同时震源技术是提高井炮激发施工效率的有效方法,但存在混叠数据分离的难题。为了解决该难题,对混叠地震数据进行稀疏变换,分析稀疏域中有效信号与混叠噪声在连续性方面的特征差异,以此为基础,开展基于广义同步压缩曲波变换和最小二乘匹配滤波的井炮同时激发混叠地震数据的分离方法研究。广义曲波变换消除常规曲波变换产生的拖尾效应,为混叠地震数据提供了精细的多方向和多尺度分解,同步压缩方法用于精准地重新分配各向异性局部波矢量的能量,广义同步曲波变换以更佳的分辨率和更高的保真度分离有效信号和邻炮干扰噪声,以最小二乘匹配滤波为正则项约束,保证在滤除噪声的同时保护有效信号。数值模拟和实际地震单炮模拟混叠数据分离测试结果表明,与矢量中值滤波方法相比,该方法能够更好地分离混叠地震数据,能较好地保护有效信号。利用缩短井炮激发时间间隔的高效激发方式在SH三维工区进行了混叠高效采集试验,混叠数据分离后的成像效果与常规逐炮激发资料的成像效果相当,验证了井炮同时激发混叠数据分离技术的可行性。研究结果表明,井炮同时震源能够大幅度提高野外放炮效率,缩短野外地震激发时间间隔,经过室内数据分离处理,得到与常规井炮逐炮激发同等质量的地震数据。
With the increasing density of shots and traces and the deployment of large three-dimensional(3-D)blocks,the workload of shooting for 3-D seismic field acquisition increases exponentially.The vibroseis high-efficiency acquisition blasts up to ten thousand shots per day,posing a challenge to the field acquisition efficiency of simultaneous shooting of borehole explosives.The core bottleneck is the separation of the simultaneous data.To solve this problem,a sparse transformation of simultaneous data was adopted to analyze the difference between the effective signal and crosstalk noise in the sparse domain.In addition,a method for separating borehole explosive simultaneous shooting data based on generalized synchronous squeezed wavelet transform and least-squares matched filtering was developed.The generalized curvilinear transform eliminated the trailing effect caused by the conventional curvilinear transform and provided fine multi-directional and multi-scale decomposition for the input-blending seismic data.Synchronous compression was used to redistribute the energy of the anisotropic local wave vector accurately.The generalized synchronous curvilinear transform separated the effective signal from the adjacent shot interference noise with better resolution and higher fidelity,with the least-squares matched filter as a regular constraint to protect the effective signal while filtering noise.The separation test for a single shot of simultaneous shooting data showed that,compared with the vector median filtering method,this method better separates mixed data and keeps the effective signal from being damaged.Seismic blending data separation technology was also used for real seismic data,and the results showed that the technique could separate the blended seismic data and maintain the effective signal.The high-efficiency acquisition of blending data was performed in SH 3D by shortening the shooting interval,and single-shot separation and profile imaging of the borehole explosive simultaneous shooting data were performed.The results were equivalent to the conventional shot-by-shot excitation imaging effect.This research proves the feasibility of mixing data separation technology,which can greatly improve the efficiency of field shooting,shorten the time interval for seismic excitation,and obtain seismic data of the same quality as the conventional method.
作者
徐雷良
邸志欣
石太昆
段伟伟
黄亮
丁建强
魏冰
XU Leiliang;DI Zhixin;SHI Taikun;DUAN Weiwei;HUANG Liang;DING Jianqiang;WEI Bing(R&D Center,Sinopec Geophysical Corporation,Nanjing 211100,China;Geological Survey and Digital Earth(Beijing)Technology Co.,Ltd,Beijing 100096,China;Institute of Geology and Geophysics,Chinese Academy of Sciences,Beijing 100029,China;University of Chinese Academy of Sciences,Beijing 100049,China)
出处
《石油物探》
CSCD
北大核心
2023年第2期281-296,共16页
Geophysical Prospecting For Petroleum
基金
中国石油化工股份有限公司科技部项目(JP21003)
中石化石油工程技术服务有限公司科技项目(SG20-62K)共同资助。
关键词
井炮
同时激发
高效地震采集
广义同步压缩曲波变换
混叠数据分离
成像
borehole explosive source
simultaneous shooting
high-efficiency seismic acquisition
generalized synchrosqueezed wavelet transform
blending data separation
imaging