地震叠后和叠前混合驱动下的页岩油储层多尺度裂缝预测方法  

Multi-scale fracture prediction of shale oil reservoir driven by the combination of post-stack and pre-stack seismic data

作　　者：陈刚[1,2] 李世昌[1] 宋斯宇 李映艳 何吉祥[2] 唐廷明 朱福玉 赵杨 卢明辉 CHEN Gang;LI ShiChang;SONG SiYu;LI YingYan;HE JiXiang;TANG TingMing;ZHU FuYu;ZHAO Yang;LU MingHui(Unconventional Oil and Gas Science and Technology Research Institute,China University of Petroleum(Beijing),Beijing 102249,China;Exploration and Development Research Institution,PetroChina Xinjiang Oilfield Company,Karamay Xinjiang 834000,China;Research Institution of PetroChina Exploration and Development,Beijing 100083,China)

机构地区：[1]非常规油气科学技术研究院,中国石油大学(北京),北京102249 [2]中国石油新疆油田公司勘探开发研究院,新疆克拉玛依834000 [3]中国石油勘探开发研究院,北京100083

出　　处：《地球物理学报》2024年第7期2830-2849,共20页Chinese Journal of Geophysics

基　　金：中国石油股份公司重大科技专项“陆相中高成熟度页岩油勘探开发关键技术研究及应用”(2019E-26);“页岩油国家级示范区水平井效益开发关键工程技术集成与试验”(2020F-50);“陆相页岩油规模增储上产与勘探开发技术研究”(2023ZZ15);中国石油股份前瞻性基础性重大科技项目“页岩油勘探开发理论技术研究”课题3“页岩油储层甜点地球物理关键参数优选与评价预测研究”(2021DJ1803)联合资助。

摘　　要：不同尺度、不同类型的天然裂缝的系统性表征有利于识别优质储层、改进开发效果.建立可靠的多尺度裂缝模型是提高页岩油储层钻遇率的重要依据,其关键在于精细预测小尺度裂缝的空间分布,指导钻井钻进和压裂过程.然而,基于地震属性的传统裂缝预测方法只能各自突出单一尺度的断裂,不同裂缝属性的数学独立性给综合裂缝系统的识别带来严峻挑战.我们提出一套改进的全尺度裂缝系统评估方案,其关键在于通过优化的各向异性蚁群裂缝识别算法确定小尺度裂缝的发育概率,将小尺度裂缝的发育概率与传统蚂蚁追踪产生的大尺度断层的概率进行融合,得到的综合裂缝系统可以实现全尺度裂缝的感知.详细的属性分析过程表明,方位各向异性能够精确描述微观裂缝特征.基于此,我们将各向异性分析得到的裂缝密度和裂缝走向融入蚁群追踪裂缝的各个阶段,在提高裂缝密集区搜索强度的同时,还对小断距裂缝的发育方向实现了精准刻画.针对页岩油藏的应用案例证明,根据改进方案得到的全尺度裂缝系统能够同时表征大尺度断层、小裂缝和裂缝破碎带的分布情况.研究区的成像测井结果进一步验证了小尺度裂缝预测发育方向的准确性及可靠性,能够为页岩油水平井开发部署提供理论指导.Systematic characterization of natural fractures across various scales and types plays a pivotal role in identifying high-quality reservoirs and optimizing development outcomes.Establishing a reliable multi-scale fracture model serves as a critical foundation for enhancing the drilling efficiency in shale oil reservoirs.The key challenge lies in accurately predicting the spatial distribution of small-scale fractures and effectively guiding drilling and fracturing operations.Traditionally,fracture prediction methods based on seismic attributes tend to emphasize individual single-scale fractures.However,the mathematical independence of different fracture attributes poses a significant obstacle when identifying integrated fracture systems.To address this,we propose an improved approach for estimating full-scale fracture systems.The key innovation involves determining the development probability of small-scale fractures using an optimized anisotropic ant colony fracture identification algorithm.By integrating this small-scale fracture probability with the likelihood of large-scale faults generated through conventional ant tracking,our method achieves comprehensive full-scale fracture sensing.The detailed property analysis process shows that the azimuthal anisotropy can accurately characterize microscopic fractures.Based on this,we integrate the fracture density and fracture orientation obtained from the anisotropy analysis into each stage of ant colony tracing fractures,which can improve the search intensity of fracture dense areas while achieving accurate portrayal of the development direction of small fracture breaks.The application cases for shale reservoirs demonstrate that the full-scale fracture system obtained according to the improved scheme can simultaneously characterize the distribution of large-scale faults,small fractures and fracture fragmentation zones.The imaging logging in the study area further validate the accuracy and reliability of predicting the direction of small-scale fractures,which can provi

关 键 词：天然裂缝 页岩油储层 方位各向异性 多尺度裂缝 蚂蚁追踪 

分 类 号：P631[天文地球—地质矿产勘探]