准噶尔盆地白家海凸起侏罗系西山窑组煤岩气“甜点”储层智能综合预测技术  

Intelligent comprehensive prediction technology of coalbed methane“sweet spot”reservoir of Jurassic Xishanyao Formation in Baijiahai uplift,Junggar Basin

作　　者：李道清[1] 陈永波[2] 杨东 李啸[1] 苏航[1] 周俊峰[2] 仇庭聪 石小茜 LI Daoqing;CHEN Yongbo;YANG Dong;LI Xiao;SU Hang;ZHOU Junfeng;QIU Tingcong;SHI Xiaoqian(Research Institute of Exploration and Development,PetroChina Xinjiang Oilfield Company,Karamay,Xinjiang 834000,China;PetroChina Research Institute of Exploration and Development-Northwest,Lanzhou 730020,China;Downhole Services company BHDC,Renqiu 062552,Hebei,China)

机构地区：[1]中国石油新疆油田公司勘探开发研究院,新疆克拉玛依834000 [2]中国石油勘探开发研究院西北分院,兰州730020 [3]中国石油集团渤海钻探工程有限公司井下作业分公司,河北任丘062552

出　　处：《岩性油气藏》2024年第6期23-35,共13页Lithologic Reservoirs

基　　金：中国石油天然气股份有限公司前瞻性基础性重大科技专项“叠合盆地中下组合规模圈闭形成机制与有效性研究”(编号:2023ZZ0205)、“深地煤岩气开发优化设计关键技术研究”(编号:2023ZZ18YJ04)联合资助。

摘　　要：为了解决准噶尔盆地白家海凸起侏罗系西山窑组煤岩厚度小、气源断裂垂向断距小导致的地震资料信噪比低及煤岩气“甜点”储层预测难度大的问题,提出了“五步法”逐级控制的测井-地质-地震一体化智能综合预测方法。研究结果表明:①“五步法”是利用构造保边去噪和谐波高频恢复处理技术提高叠前CRP道集的信噪比和分辨率;通过调谐厚度法和分频智能反演法相结合定量预测煤岩厚度分布;利用深度学习智能断裂检测技术预测气源断裂展布特征;基于煤岩流体替换对不同含气饱和度时的AVO特征进行分析,通过含气饱和度预测含气分布范围;采用叠合分析法预测“甜点”储层,即位于断鼻(或断块)、煤岩厚度大,存在气源断裂及含气饱和度高于50%的叠合部位。②研究区“甜点”主要分布在北部走滑断裂的南北两侧断鼻或断块圈闭中,共发育31个煤岩气“甜点”区,累计面积达231.9 km^(2),其中走滑断裂北侧的5个“甜点”储层勘探潜力更大。③依据“五步法”部署的预探井与实钻井考核指标的吻合率达92%;对部署的水平井井轨迹进行了优化设计和动态监控,提高了单井产能。In order to solve the problems of small coal thickness,small vertical fault distance of gas source fault,low signal-to-noise ratio of seismic data and difficult to predict the sweet spot reservoir of coal,rock and gas in Baijiahai uplift,Junggar basin.A“five-step method”is proposed,which is controlled stepwise.The results show that:(1)“Five-step”comprehensive technology provides a powerful technical means for coalbed methane“sweet spot”reservoir prediction,The specific method is to improve the signal-to-noise ratio and resolution of CRP gathers using construction edge preserving denoising and harmonic high-frequency recovery processing techniques;The combination of tuning thickness method and frequency division intelligent inversion method,the thickness distribution range of coalbed section and plane is quantitatively predicted;The deep learning intelligent fracture detection technology was used to predict the fracture profile and plane distribution characteristics of gas source;Analysis of AVO characteristics with different gas saturation based on fluid replacement of coal and rock,the gas saturation is used to predict the gas distribution range in the study area;The“sweet spot”reservoir is located in the superposition of broken nose(or fault block),large coal rock thickness,gas source fracture and high gas saturation.(2)The“sweet spot”is mainly distributed in the fault nose or block trap on the north and south sides of the strike-slip fault in the northern part of the work area,A total of 31 coalbed methane“sweet spot”regions have developed,with a cumulative area of 231.9 km^(2),The exploration potential of the five“sweet spot”reservoirs on the north side of the strike-slip fault is greater.(3)The coincidence rate between the test index of the vertical well deployed by the research results and the real drilling results is 92%.At the same time,the trajectory of the horizontal well can be optimized and dynamically monitored to improve the productivity of a single well.

关 键 词：煤岩气 叠前CRP道集处理 分频智能反演 智能断裂检测 深度学习 “甜点”预测 西山窑组 侏罗系 白家海凸起 准噶尔盆地 

分 类 号：TE122[石油与天然气工程—油气勘探] P618.13[天文地球—矿床学]