时空-频率波数域同步提取变换及其在致密河道砂岩储层表征中的应用  

Space-time frequency wavenumber domain synchroextracting transform and its application to tight channel sandstone reservoir characterization

作　　者：方月 陈辉[1,2,3] 王思媛 陈旭平 兰萍 FANG Yue;CHEN Hui;WANG SiYuan;CHEN XuPing;LAN Ping(State Key Laboratory of Geological Hazard Prevention and Geological Environment Protection(Chengdu University of Technology),Chengdu 610059,China;Sichuan Provincial Key Laboratory of Mathematical Geology,Chengdu University of Technology,Chengdu 610059,China;School of Mathematical Sciences,Chengdu University of Technology,Chengdu 610059,China)

机构地区：[1]地质灾害防治与地质环境保护全国重点实验室(成都理工大学),成都610059 [2]数学地质四川省重点实验室(成都理工大学),成都610059 [3]成都理工大学数学科学学院,成都610059

出　　处：《地球物理学报》2025年第3期1163-1173,共11页Chinese Journal of Geophysics

基　　金：国家自然科学基金(42174164,41704132);四川省科技教育联合基金项目(2024NSFSC1955);四川省自然科学基金项目(2024NSFSC0080,2023NSFSC5116);地质灾害防治与地质环境保护国家重点实验室自主研究课题(SKLGP2022Z011);成都理工大学研究生拔尖创新人才培育项目“同步挤压/提取变换域致密河道砂岩气储层地震信号时空表征方法”(CDUT2023BJCX002);微分方程四川省自然科学基金创新研究群体(2023NSFSC1984);四川省科技计划项目(2021YJ0358,2019YJ0402);中国海洋石油有限公司“十四五”重大科技项目(KJGG2022-0903);教育部“春晖”计划项目联合资助。

摘　　要：致密河道砂岩储层作为非常规油气领域的重点勘探开发对象,其形态特征可以在频谱分析中突出显示.然而,现有时频分析方法大多侧重于对地震数据进行逐道分析,缺乏横向约束,不能详细描绘出河道的形态特征.为此,本文提出一种时空-频率波数域同步提取变换(Space-Time Frequency Wavenumber Domain Synchroextracting Transform,STFKSET).该方法引入一个时空窗口来捕捉地震信号的横向变化,然后从二维纯谐波信号出发,在二维短时傅里叶变换(Two-dimensional Short-Time Fourier Transform,TDSTFT)时频域上准确推导时空-瞬时频率(SpaceTime Instantaneous Frequency,ST-IF)和时空-瞬时波数(Space-Time Instantaneous Wavenumber,ST-IK)的估计式,并定义一个新的时空-频率波数域同步提取算子(Space-Time Frequency Wavenumber Domain Synchroextracting Operator,STFKSEO),仅提取与ST-IF和ST-IK曲线高度相关的时频结果系数,以此提升对于噪声的鲁棒性,实现信号时频表征的高度聚焦性.最后构造出能有效反映储层纵横向变化的STFKSET,并根据ST-IF与ST-IK的关系来确定波数参数.合成地震剖面表明,STFKSET能够进行精细的频率脊线提取,同时保持较高的横向连续性,STFKSET具有优于传统时频分析方法的抗噪性能;将STFKSET应用到中国四川盆地西部某气田的实际地震数据处理中,进一步验证了该时频算法在刻画河道及其边界的潜力.As a key exploration and development object in the field of unconventional oil and gas,the tight channel sandstone reservoir can be highlighted in the spectrum analysis.However,most of the existing timefrequency analysis methods focus on the trace-by-trace analysis of seismic data,lack of lateral constraints,and fail to detail the morphological characteristics of the river.In this paper,a Space-Time Frequency Wavenumber Domain Synchroextracting Transform(STFKSET)is proposed.The method introduces a time-space window to capture the lateral variations in seismic signals,and then starts from two-dimensional pure harmonic signals,the Two-Dimensional Short-Time Fourier Transform(TDSTFT),The estimators of Space-Time Instantaneous Frequency(ST-IF)and Space-Time Instantaneous Wavenumber(ST-IK)are accurately derived from TDSTFT in time-frequency domain.A new Space-Time Frequency Wavenumber Domain Synchroextracting Operator(STFKSEO)was defined to extract only time-frequency result coefficients highly correlated with ST-IF and STIK curves,so as to improve the robustness to noise.Achieve high focus of signal time-frequency characterization.Finally,STFKSET which can effectively reflect the vertical and horizontal changes of the reservoir is constructed,and the wave number parameters are determined according to the relationship between ST-IF and ST-IK.The synthetic seismic section shows that STFKSET can extract frequency ridges in fine detail while maintaining high transverse continuity.STFKSET has better anti-noise performance than traditional timefrequency analysis methods.STFKSET is applied to the seismic data processing of a gas field in the western Sichuan Basin,China,and the potential of the time-frequency algorithm in the characterization of river channels and their boundaries is further verified.

关 键 词：时频分析 时空-频率波数域同步提取变换 横向连续 致密河道砂岩储层 边界刻画 

分 类 号：P631[天文地球—地质矿产勘探]