摘要
碎屑岩储层是中国乃至全球油气资源的重要载体。受限于碎屑岩储层非均质强、地下表征资料相对不足的客观条件,传统的表征与建模技术长期以来难以满足地下储层精细勘探与高效开发的需求。21世纪以来,众多学者逐步尝试将人工智能技术引入碎屑岩储层表征与建模领域,并在近10年取得快速发展。鉴于此,作者系统梳理了智能化技术在碎屑岩储层表征与建模领域的发展历程与研究现状,重点阐述了储层参数测井智能解释、智能化断层与地层构造解析、井震融合智能储层预测、碎屑岩储层智能三维地质建模的最新研究进展与应用效果,并分析了不同智能化储层表征与建模技术面临的挑战与未来的发展方向。总体而言,上述碎屑岩储层智能表征技术在不同程度上面临着高质量样本不足、智能学习模型泛化能力较差、知识驱动与数据驱动融合程度低等难题,未来仍有巨大的发展空间与良好的应用前景。
Clastic rock reservoirs serve as critical carriers of hydrocarbon resources both in China and around the world.However,due to inherent limitations such as strong heterogeneity and insufficient subsurface characterization data,traditional methods of reservoir characterization and modeling have struggled to fulfill the demands for high-resolution exploration and efficient development.Since the 21th century,researchers have progressively integrated artificial intelligence(AI)techniques into the field of clastic reservoir characterization and modeling,resulting in significant advancements over the past decade.These innovations have significantly improved both the accuracy and efficiency of reservoir characterization.In this context,this paper systematically reviews the development history and current research status of intelligent technologies in clastic reservoir characterization and modeling.It highlights recent progress and application outcomes in areas such as intelligent well log interpretation for reservoir parameters,AI-based fault and stratigraphic framework analysis,intelligent reservoir prediction through well-seismic integration,and intelligent 3D geological modeling.Furthermore,we discuss the challenges faced by various intelligent approaches and outlines future directions for their development.Overall,these intelligent characterization techniques have made significant advances and demonstrated positive outcomes in practical applications.Nevertheless,they also face multiple challenges,including a lack of high-quality training samples,suboptimal generalization capabilities of learning models,and inadequate coupling of knowledge-driven with data-driven approaches.Despite these limitations,there remains significant potential for advancement,with promising application prospects emerging across reservoir characterization workflows.
作者
岳大力
李伟
王武荣
孙盼科
吴胜和
徐振华
刘磊
邬德刚
屈林博
任柯宇
林津
张姝琪
YUE Dali;LI Wei;WANG Wurong;SUN Panke;WU Shenghe;XU Zhenhua;LIU Lei;WU Degang;QU Linbo;REN Keyu;LIN Jin;ZHANG Shuqi(State Key Laboratory of Petroleum Resources and Engineering,China University of Petroleum(Beijing),Beijing 102249,China;College of Geosciences,China University of Petroleum(Beijing),Beijing 102249,China;College of Artificial Intelligence,China University of Petroleum(Beijing),Beijing 102249,China)
出处
《古地理学报》
北大核心
2025年第4期903-923,共21页
Journal of Palaeogeography:Chinese Edition
基金
国家自然科学基金项目(编号:42272186,42202109,42302128,42412179)资助。
关键词
碎屑岩
储层表征
三维地质建模
测井解释
井震融合
人工智能
clastic rock
reservoir characterization
3D geological modeling
well-log interpretation
well-seismic integration
artificial intelligence
