Structure analysis and intelligent prediction of carbonate fractured-vuggy reservoirs in ultra-deep fracture zone

doi:10.13745/j.esf.sf.2024.5.31

Abstract

Abstract:

Ultra-deep carbonate fractured-vuggy reservoirs, typically located near fault zones, are characterized by significant burial depth, large vertical extent, and narrow horizontal distribution, making seismic identification highly challenging. Conventional fracture-vuggy identification techniques primarily rely on amplitude response characteristics, which only provide the general orientation and approximate extent of the reservoir. To address these limitations, a refined interpretation method for fractured-vuggy reservoirs was proposed based on a two-dimensional Res-UNet residual network and optimized label data. This method enhances the update frequency and optimization efficiency of 2-D training samples by modifying the Res-UNet residual network structure, thereby improving the accuracy of reservoir prediction. Geological models of faults, fractures, and karst caves were constructed using a combination of satellite imagery, UAV scanning, field reconnaissance, geological radar, and seismic interpretation results. Wavelet analysis of 3-D seismic data in the depth domain, along with imaging and acoustic logging, was performed to extract formation and reservoir velocities. Wavelets at frequencies of 20 Hz, 25 Hz, and 35 Hz were selected to create forward models for different reservoir widths. An empirical formula was derived to establish the relationship between reservoir characteristics and seismic response based on frequency and formation velocity. Verification of the empirical formula was conducted using horizontal well trajectories, vertical well remote detection data, and logging data to extract the relationship between actual reservoir characteristics and seismic responses. This formula was then applied to amplitude attributes of the target layer in the study area to estimate the reservoir width range. The reservoir combination characteristics of the study area were identified, and a comprehensive detection workflow—incorporating analysis, design, verification, and redesign—was established through the analysis of 3-D seismic profiles. Optimal parameters, including wavelet frequency, migration aperture, and sampling interval, were designed for the forward model of the study area. A 2-D training sample database capable of real-time updates was developed, and the fracture-vuggy labels were iteratively optimized by integrating synthetic fracture-vuggy data with actual data. Using this database, the Res-UNet residual network was trained and applied to address the degradation problem commonly encountered in deep networks. This approach enabled the fine-scale prediction of fracture-vuggy structures, significantly improving the resolution and accuracy of reservoir interpretation.

Key words: Tarim Basin, fault-controlled fracture-vuggy reservoirs, deep learning, Res-UNet residual network, sample library optimization

CLC Number:

P631

LI Fenglei, LIN Chengyan, WANG Jiao, REN Lihua, ZHANG Guoyin, ZHU Yongfeng, LI Shiyin, ZHANG Yintao, GUAN Baozhu. Structure analysis and intelligent prediction of carbonate fractured-vuggy reservoirs in ultra-deep fracture zone[J]. Earth Science Frontiers, 2025, 32(2): 311-331.

Figures/Tables 25

Fig.1 The regional geological background of Tarim Basin and the structural characteristics of the study area

Fig.2 The geological model of fracture-vuggy body is established based on geological outcrop and drilling information

Fig.3 Seismic response and single well production of typical well group fracture-vuggy body in the study area

Fig.4 The corresponding relationship between FMI and AC of various types of reservoirs

Fig.5 The analysis of velocity curve and distribution range of strata and reservoirs in the study area based on AC

Fig.6 The relationship between time domain and depth domain Rake wavelet

Fig.7 Characteristics of Rake wavelet profile in depth domain

Fig.8 The results obtained by migration calculation with different migration apertures

Fig.9 The logging curve and seismic profile corresponding to the drilling trajectory of the inclined fracture zone

Fig.10 The results of acoustic remote detection of reservoir in fault fracture zone and its seismic data characteristics

Fig.11 Seismic forward results and the relationship between the width of reservoirs and its seismic response

Fig.12 Reservoirs and its seismic response empirical formula fitting process diagram

Table 1 The seismic response width corresponding to the actual drilling results is compared with the formula calculation results

储层实际宽度/m	计算串珠宽度/m	实测串珠宽度/m	误差/%
35	145.6	150	-2.9
50	175.5	180	-2.5
120	258.0	250	3.2

Table 2 The change of oil column height in the study area based on the change of crude oil production temperature

油柱高度/m	油水界面/m	压降	产量	见水周期/月	井底流温变化	参与分析井
>500	300~500	稳定	高	尚未见水	Yuem5,7.2 ℃	Yuem5、Yuem2-3x、Yuem10、Yuem2-1、Yuem2-5x、 Yuem2-H6和Yuem2-H7
>200	100~200	见水波动	中等	30~50	无数据	Yuem5-1、Yuem5-3、Yuem5-5和Yuem4-4X
>100	50~100	快、不稳	低	1~20	无数据	Yuem2、Yuem2-2、Yuem5-2和Yuem5-4X

Fig.13 The forward model and prediction results of typical section interpretation results

Fig.14 Comparison of different module structures

Fig.15 The residual module structure used in this paper

Fig.16 two-dimensional residual Unet coding-decoding network of fractured-vuggy reservoirs

Fig.17 Geological model guides the label optimization process

Fig.18 High-precision two-dimensional fracture-vuggy fabrication and label optimization flow chart

Fig.19 The accuracy and loss of the network training iteration 300 times are obtained

Fig.20 The fracture-vuggy results predicted by network training

Fig.21 The seismic data of different work areas are compared by sample difference

Fig.22 Comparison of seismic data characteristics and prediction results in the study area

Fig.23 The prediction results are compared with other reservoir prediction results in this paper

References 51

[1]	ZHAO W Z, SHEN A J, QIAO Z F, et al. Carbonate Karst reservoirs of the Tarim basin, Northwest China: types, features, origins, and implications for hydrocarbon exploration[J]. Interpretation, 2014, 2(3):65-90.
[2]	ZHU G Y, MILKOV A V, ZHANG Z Y, et al. Formation and preservation of a giant petroleum accumulation in superdeep carbonate reservoirs in the southern Halahatang oil field area, Tarim basin, China[J]. Bulletin of the American Association of Petroleum Geologists, 2019, 103(7): 1703-1743.
[3]	樊太亮, 高志前, 吴俊. 塔里木盆地深层碳酸盐岩建造-改造作用与多类型储层有序性分布[J]. 地学前缘, 2023, 30(4): 1-18. DOI
[4]	李凤磊, 林承焰, 任丽华, 等. 塔里木盆地塔北地区多期断裂差异匹配控制下超深岩溶缝洞储层成藏特征[J/OL]. 地学前缘, 1-18[2024-2-05]. https://doi.org/10.13745/j.esf.sf.2023.9.5.
[5]	KREY T. The significance of diffraction in the investigation of faults[J]. Geophysics, 1952, 17(4):843-858. DOI: 10.1190/1.1437815.
[6]	赵惊涛, 王真理, 于彩霞. 地震勘探中的边缘绕射波及其动力学识别方法[J]. 地球物理学进展, 2011, 26(1): 194-206.
[7]	李阳. 塔河油田奥陶系碳酸盐岩溶洞型储集体识别及定量表征[J]. 中国石油大学学报(自然科学版), 2012, 36(1): 1-7.
[8]	李鹏飞, 崔德育, 黄诚. 地震资料处理解释一体化技术在塔北碳酸盐岩储层识别中的应用[J]. 石油地球物理勘探, 2018, 53(增刊2): 306-313, 19.
[9]	徐红霞, 沈春光, 李斌, 等. 多属性分析技术在碳酸盐岩断溶体预测中的应用[J]. 石油地球物理勘探, 2017, 52(增刊2): 158-163, 8.
[10]	鲁新便, 胡文格, 汪彦. 塔河地区碳酸盐岩断溶体油藏特征与开发实践[J]. 石油与天然气地质, 2015, 36(3): 347-355.
[11]	姜仁旗, 吴键, CASTAGNA J, 等. 地球物理技术最新进展: 高分辨率地震频率和相位属性分析技术研究与应用效果[J]. 地学前缘, 2023, 30(1): 199-212. DOI
[12]	杨午阳, 杨佳润, 陈双全, 等. 基于U-Net深度学习网络的地震数据断层检测[J]. 石油地球物理勘探, 2021, 56(4): 688-697, 669.
[13]	WANG J, ZHANG J H, ZHANG J L, et al. Research on fault recognition method combining 3D Res-UNet and knowledge distillation[J]. Applied Geophysics, 2021, 18(2): 199-212.
[14]	王康, 刘彩云, 熊杰, 等. 基于全卷积残差收缩网络的地震波阻抗反演[J]. 物探与化探, 2023, 47(6): 1538-1546.
[15]	王莉利, 杜功鑫, 石颖, 等. 基于MultiResAttUnet网络的二维地震断层自动识别方法研究[J]. 地球物理学进展, 2023, 38(5): 2160-2171.
[16]	王玥天, 毛志强, 胡琮, 等. 基于深度卷积神经网络的岩石矿物组分含量测井评价方法研究[J]. 地球物理学进展, 2023, 38(2): 748-758.
[17]	孙冲, 雷刚林, 张银涛, 等. 基于深度残差网络的走滑断层智能识别方法: 以塔里木盆地富满油田为例[J]. 石油物探, 2024, 63(1): 129-137. DOI
[18]	张文起, 李春雷. 基于多层残差网络的地震提频处理在薄储集层识别中的应用[J]. 新疆石油地质, 2024, 45(1): 102-108.
[19]	WU X M, YAN S S, QI J, et al. Deep learning for characterizing paleokarst collapse features in 3-D seismic images[J]. Journal of Geophysical Research: Solid Earth, 2020, 125(9): e2020JB019685.
[20]	ZHANG G Y, LIN C Y, REN L H, et al. Seismic characterization of deeply buried paleocaves based on Bayesian deep learning[J]. Journal of Natural Gas Science and Engineering, 2022, 97: 104340.
[21]	闫星宇, 李宗杰, 顾汉明, 等. 基于深度卷积神经网络的地震数据溶洞识别[J]. 石油地球物理勘探, 2022, 57(1): 1-11, 264.
[22]	DECKER L, JANSON X, FOMEL S. Carbonate reservoir characterization using seismic diffraction imaging[J]. Interpretation, 2015, 3 (1): SF21-SF30.
[23]	余为维, 冯磊, 杜艳艳, 等. 煤层陷落柱的三维地震槽波正演模拟[J]. 煤炭技术, 2019, 38(3): 73-76.
[24]	王勤耕, 胡善政, 黄有晖, 等. 龙门山北段逆掩推覆构造三维地球物理模型构建[J]. 石油地球物理勘探, 2023, 58(6): 1359-1364.
[25]	姚姚, 唐文榜. 深层碳酸盐岩岩溶风化壳洞缝型油气藏可检测性的理论研究[J]. 石油地球物理勘探, 2003, 38(6): 623-629, 708-579.
[26]	杨平, 孙赞东, 李海银, 等. 影响岩溶缝洞体地震反射特征的关键因素分析[J]. 石油地球物理勘探, 2015, 50(3): 523-529, 6.
[27]	李凡异, 魏建新, 狄帮让, 等. 碳酸盐岩孔洞储层地震物理模型研究[J]. 石油地球物理勘探, 2016, 51(2): 272-280, 205.
[28]	张军华, 朱文博, 吴成, 等. 碳酸盐岩溶洞成像要素分析与研究[J]. CT理论与应用研究, 2014, 23(3): 413-423.
[29]	刘宝增, 漆立新, 李宗杰, 等. 顺北地区超深层断溶体储层空间雕刻及量化描述技术[J]. 石油学报, 2020, 41(4): 412-420. DOI
[30]	郑多明, 李志华, 赵宽志, 等. 塔里木油田奥陶系碳酸盐岩缝洞储层的定量地震描述[J]. 中国石油勘探, 2011, 16(增刊1): 57-62, 78, 172-173.
[31]	李阳, 侯加根, 李永强. 碳酸盐岩缝洞型储集体特征及分类分级地质建模[J]. 石油勘探与开发, 2016, 43(4): 600-606. DOI
[32]	杜赫, 徐守余, 冯建伟, 等. 基于数字露头表征的岩溶缝洞组构特征[J]. 中国石油大学学报(自然科学版), 2020, 44(5): 1-9.
[33]	LOUCKS R G. Modern analogs for paleocave sediment fills and their importance in identifying paleocave reservoirs[J]. AAPG Bulletin, 2001, 51:195-206.
[34]	司文朋, 薛诗桂, 马灵伟, 等. 顺北走滑断裂-断溶体物理模拟及地震响应特征分析[J]. 石油物探, 2019, 58(6): 911-919.
[35]	张银涛, 陈石, 刘强, 等. 塔里木盆地富满油田FⅠ19断裂发育特征及演化模式[J]. 现代地质, 2023, 37(2): 283-295.
[36]	张银涛, 余一欣, 谢舟, 等. 塔里木盆地富满地区走滑断裂带精细刻画及勘探应用成效[J]. 现代地质, 2024, 38(6): 1417-1430.
[37]	李凤磊, 林承焰, 崔仕提, 等. 塔北地区奥陶系古地貌及走滑断裂差异性控储规律[J]. 中国石油大学学报(自然科学版), 2022, 46(6): 48-58.
[38]	WU G H, MA B S, HAN J F, et al. Origin and growth mechanisms of strike-slip faults in the central Tarim cratonic basin, NW China[J]. Petroleum Exploration and Development, 2021, 48(3): 595-607. DOI
[39]	张军华, 张彬彬, 吴成, 等. 地震采集与处理因素对溶洞成像的影响分析[J]. 地球物理学进展, 2014, 29(3): 1350-1356.
[40]	温声明, 杨德兴, 王贵重, 等. 层速度替换建场法在高陡构造区的应用[J]. 石油地球物理勘探, 2006, 41(3): 308-312, 362, 17.
[41]	张蓉, 徐群洲, 帕尔哈提, 等. 对提高地层压力预测精度的探讨: 声波测井层速度与地震层速度的关系[J]. 石油天然气学报, 2010, 32(2): 274-276.
[42]	张滨鑫, 张冠杰, 刘敬寿, 等. 碳酸盐岩缝洞型储层识别与雕刻方法研究进展[J]. 地球物理学进展, 2024, 39(2): 580-593.
[43]	马勇, 黎雨航, 张卫峰, 等. 基于成像测井的塔河油田奥陶系鹰山组中深层碳酸盐岩优质储集层预测[J]. 地球物理学进展, 2023, 38(6): 2525-2539.
[44]	张莹莹, 王云专, 石颖, 等. 雷克子波频率研究[J]. 地球物理学进展, 2017, 32(5): 2162-2167.
[45]	WANG Y H. Frequencies of the ricker wavelet[J]. Geophysics, 2015, 80(2): A31-A37.
[46]	李雪英, 程云, 聂伟东, 等. 基于两步扫描法的Kirchhoff时间最优偏移孔径[J]. 东北石油大学学报, 2019, 43(3): 12-19, 5-6.
[47]	岳玉波, 李振春, 钱忠平, 等. 转换波Kirchhoff叠前时间偏移的成像优化方案[J]. 地球物理学报, 2018, 61(3): 1188-1195. DOI
[48]	许孝凯. 声波远探测测井与地震缝洞精细评价技术[J]. 应用声学, 2023, 42(6): 1185-1191.
[49]	张承森, 肖承文, 刘兴礼, 等. 远探测声波测井在缝洞型碳酸盐岩储集层评价中的应用[J]. 新疆石油地质, 2011, 32(3): 325-328.
[50]	He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]. Las Vegas, NV:2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016: 770-778.
[51]	李凤磊, 林承焰, 张国印, 等. 塔北地区多期走滑断裂地球物理响应特征及精细识别[J]. 中国石油大学学报(自然科学版), 2024, 48(3): 1-14.