走向精确地震勘探的道路

2021年 60卷 第No. 5期
阅读:99
查看详情
Pathway toward accurate seismic exploration
(波现象与智能反演成像研究组(WPI),同济大学海洋与地球科学学院,上海200092)
(Wave Phenomena and Intelligent Inversion Imaging Group,School of Ocean and Earth Science,Tongji University,Shanghai 200092,China)
勘探地震领域已扩展到复杂地表、复杂构造、复杂油藏和深层目标等。地震数据采集技术发展到了宽方位、高密度、宽频带(“两宽一
高”)地震数据采集阶段;地震波成像技术发展到了贝叶斯(Bayes)参数估计理论下的全波形反演(FWI)和最小二乘逆时深度偏移成像
(LS_RTM);油藏描述发展到了综合信息利用和最佳判定阶段;地震勘探技术已经发展至全新的阶段。横向缓变的层状介质假设、地表一致性假
设、射线理论波传播和Zoeppritz方程界定了上一代精确地震勘探的方法技术及其适用性,上一代精确地震勘探以高分辨率地震子波作为成像
处理的核心目标,并据此开展薄层油气藏的识别、描述与评价。而描述任意介质中地震波传播的波动理论和贝叶斯参数估计理论构成了新一
代高精度地震勘探的理论基础。“两宽一高”的地震数据采集技术和更高精度的子波处理;基于高维、字典基和稀疏特征表达的信号处理技
术(解决去噪、数据规则化、数据压缩、去混叠等问题)、建立更精确速度和Q值模型以及估计宽带反射系数的特征波反演成像技术、宽带波
阻抗成像技术和基于信息综合的人工智能油藏描述技术代表了走向精确地震勘探的未来方向。

The domain of seismic exploration has expanded to cover complex topographies,complex subsurface structures,complex reservoirs,and deep geological targets.The new generation of accurate seismic exploration methods entail wide-azimuth,wide-frequency band,and high-density seismic data acquisition.In these methods,full-wave inversion,and least-square reverse time migration methods based on Bayes estimation theory have become the standard for imaging,and reservoir characterization is implemented with integrated information utilization and optimal inference.Evidently,seismic exploration has entered a new era.For the last-generation accurate seismic exploration,the assumption of consistence for layers and surfaces,the ray-theory wave propagation,and Zoeppritz equations define the scope of seismic data processing and its applicability.A high-resolution seismic wavelet is the key objective of seismic data acquisition and imaging methods,based on which the identification,characterization,and evaluation of thin reservoirs is carried out.In new-generation methods,the theoretical basis is enriched by a number of tools,namely:the Bayesian estimation theory and wave equations that describe wave propagation in complex media.The seismic data acquisition with wide-azimuth,wide-frequency band,and high-density sampling;accurate seismic wavelet processing;amplitude and phase preserving signal processing based on high dimension,dictionary bases and sparse representation;characteristic wave inversion for accurate velocity model building;Q model building and wideband reflectivity estimation;wideband impedance model building;and reservoir characterization with integrated information utilization and artificial intelligence,all of these methods provide novel pathways for accurate seismic exploration.

“两宽一高”地震数据; 全波形反演; 特征波反演; 背景波阻抗建模; 宽带方位角度反射系数; 宽带波阻抗建模; 精确油藏描述; 机器学习与人工智能; 精确地震勘探
 
;
“wide-azimuth;; wide-frequency band;; and high-density” seismic data acquisition;; full wave inversion;; characteristic wave inversion;; background impedance model building;; wideband azimuth and opening-angle reflectivity;; wideband impedance model building;; accurate reservoir characterization;; machine learning and artificial intelligence;; accurate seismic exploration;

国家重点研发计划变革性技术关键科学问题重点专项(2018YFA0702503)、国家重点研发计划深海关键技术与装备重点专项(2019YFC0312004)、国家自然科学基金(41774126)、南方海洋科学与工程广东省实验室(湛江)资助项目(ZJW-2019-04)共同资助。

10.3969/j.issn.1000-1441.2021.05.001