基于地震属性智能融合的稀井网辫状河储层构型精细表征 --以渤海湾盆地C-6油田馆陶组为例
- 收稿日期:
2023-12-20
- 网络出版日期:
2024-03-15
摘要: C-6油田是渤海海域亿吨级曹妃甸油田群主力油田之一,其主力开发层系馆陶组Ⅲ油组为一套富砂的辫状河沉积,辫状河储砂体结构及其连通性是影响油田开发效果的关键地质因素。论文采用基于地震属性智能融合技术,在有限测井信息的标定下,对油田辫状河储层四级构型单元空间分布进行了精细表征。综合多井测井解释,C-6油田馆Ⅲ油组主要发育心滩和辫状河道两种类型四级构型单元,其中心滩储层厚度大,物性较好,是研究区主要发育的储层构型单元。在地震属性提取与单井岩性和物性参数相关性分析的基础上,选取反射强度、相对波阻抗、甜点、原始振幅、瞬时振幅5种属性基于孔隙度监督的DFNN(深度前馈神经网络)智能融合,得到了反映辫状河岩性和物性的三维融合数据体,大幅度提高了辫状河储层砂体及其边界的探测能力,能够有效地开展辫状河储层四级构型单元平面和剖面分布的精细刻画。C-6油田馆Ⅲ油组主体为一北东-南西向辫流带,内部划分出呈菱形的15个心滩四级构型单元,分流河道四级构型单元呈窄条带状环绕在心滩周围,二者之间的四级构型界面对流体运移能够起到一定渗流屏障作用,垂向上心滩互相切叠,形成“大心滩-小河道”的平面构型组合样式。基于地震属性智能融合的储层构型精细表征深化了稀井网控制的辫状河储层连通性认识,为C-6油田开发方案的调整提供了直接的地质依据。
Fine characterization of braided river reservoir architecture with sparse well pattern based on intelligent fusion of multiple seismic attributes-- A Case study of Guantao Formation of C-6 Oilfield , Bohai Bay Basin
- Received Date:
2023-12-20
- Available Online:
2024-03-15
Abstract: C-6 Oilfield is one of the most principal oilfields of Caofeidian oil province with hundred million cubic metre of reserves. The third oil group of Guantao Formation (N1gⅢ), the main production zone of C-6 Oilfield, is sand-rich braided river deposit. The architecture and connectivity of braided river sandstone are the key geological factors affecting the development effect of the Oilfield. Calibrated with limited log data, intelligent fusion technology of multiple seismic attributes was introduced to finely characterizes the spatial distribution of the level-4 architecture units of the braided river reservoir. According to log interpretation, N1gⅢ of C-6 oilfield mainly develops two types of level-4 architecture units, namely, channel bar and braided channel, and braided bar is the best reservoir with high sandstone thickness and excellent physical properties. Based on seismic attribute extraction and correlation analysis with lithological and physical parameters, reflection intensity, relative impedance, sweet point, original amplitude, envelop were chosen as intelligent fusion seismic attributes with Deep Feed-Forward Neural Network (DFNN) algorithm under the supervision of porosity. The 3D attribute of DFNN fusion, representative of lithology and petrophysical property, largely improves the detecting ability of braided river sandstone unit and its boundary, and can be used to finely characterize the plan and section distribution of braided river level-4 architecture units effectively. A NE-SW braided flow zone was developed in N1gⅢ of C-6 oilfield, which could be internally sub-divided into 15 rhombic level-4 architecture units of the braided bar. Distributary channels, another level-4 architecture units, surrounded braided bar in a narrow strip. The level-4 architecture interface between the two units played as seepage barriers for fluid migration. The braided bars cut and overlapped one another vertically, forming “big bar and small channel” plan reservoir architecture pattern. The fine characterization results deepened the understanding of the reservoir connectivity of the braided river reservoir with sparse well pattern, which provided direct geological bases for the making of optimized adjustment plan of C-6 Oilfield.
尹志军, 李彦泽, 张建民, 张章, 侯东梅, 陈冰歌. 基于地震属性智能融合的稀井网辫状河储层构型精细表征 --以渤海湾盆地C-6油田馆陶组为例[J]. 沉积学报. doi: 10.14027/j.issn.1000-0550.2024.022
Fine characterization of braided river reservoir architecture with sparse well pattern based on intelligent fusion of multiple seismic attributes-- A Case study of Guantao Formation of C-6 Oilfield , Bohai Bay Basin[J]. Acta Sedimentologica Sinica. doi: 10.14027/j.issn.1000-0550.2024.022
Citation: |
Fine characterization of braided river reservoir architecture with sparse well pattern based on intelligent fusion of multiple seismic attributes-- A Case study of Guantao Formation of C-6 Oilfield , Bohai Bay Basin[J]. Acta Sedimentologica Sinica. doi: 10.14027/j.issn.1000-0550.2024.022
|