基于宽频地震信息的薄储层预测方法研究及应用  

Research and application of thin reservoir prediction method based on broadband seismic information

作　　者：邓勇 刘啸虎 王丽丽 张大伟 朱峰 苗金鹏 尹照普 DENG Yong;LIU Xiaohu;WANG Lili;ZHANG Dawei;ZHU Feng;MIAO Jinpeng;YIN Zhaopu(Research Institute of BGP,Urumqi 830016,China)

机构地区：[1]东方地球物理勘探有限责任公司研究院乌鲁木齐分院,乌鲁木齐830016

出　　处：《物探化探计算技术》2025年第2期209-221,共13页Computing Techniques For Geophysical and Geochemical Exploration

基　　金：中国石油集团东方地球物理勘探有限责任公司科研项目(03-03-2023)。

摘　　要：主流薄储层预测方法对钻井资料依赖度较高,适用于钻井较多且分布较为均匀的开发区块,而对于勘探程度低、钻井少且分布不均匀的地区预测精度较低,难以满足岩性精细勘探的需求。笔者通过对薄储层的测井、地震频谱特征精细分析,采用连续小波变换和阻尼因子对原始地震数据合理拓频并重塑的基础上。首先,应用包含低频信息的纯波数据进行常规反演,提取低频地质信息并对常规测井模型进行修正;然后,以频谱整形的高频数据和低频反演修正的模型作为地震输入,实现地震数据高低频信息的融合,完成分频迭代反演。该方法在提高地震资料纵向分辨率的同时,充分利用地震资料的低频信息弥补高频数据反演过程中导致横向分辨率变差的问题,有效提高了薄砂体纵、横向识别精度,实现了在勘探程度低、钻井资料少地区对薄储层刻画的需求。通过在准噶尔盆地东部北43井区的应用,准确落实二叠系10 m以上储层,预测吻合率达到83%以上,能够为井位部署提供有利依据。The mainstream thin reservoir prediction method is highly dependent on drilling data and is suitable for the development block with more drilling and more uniform distribution,while the prediction accuracy is low for the area with low exploration degree,few drilling and uneven distribution,and it is challenging to meet the needs of fine lithology exploration.In this paper,through the fine analysis of logging and seismic spectral characteristics,firstly,the continuous wavelet transform and damping factor are proposed to extend frequency and reshape the original seismic data reasonably.Then,the low-frequency pure wave data is used for conventional inversion to extract low-frequency information,which is used to revise the low-frequency model of iterative inversion.Finally,the frequency-reshaped high-frequency data and the low-frequency inversion modified model are used as seismic input to realize the fusion of high and low-frequency information of seismic data and complete the frequency-division iterative inversion.This method not only improves the vertical resolution of seismic data,but also makes full use of the low-frequency information of seismic data to make up for the problem of the poor lateral resolution caused by the inversion of high-frequency data,effectively improving the vertical and horizontal identification accuracy of thin sand body,and realizes the demand of thin reservoir characterization in areas with low exploration degree and few drilling data.This method predicts the over 10 meters’thick reservoir in the Permian accurately in Bei43 well area in the eastern junggar basin,the predicted agreement rate is more than 83%and provide valuable support for the well proposal.

关 键 词：薄储层预测 资料拓频 频谱整形 分频迭代反演 岩性勘探 

分 类 号：P618.13[天文地球—矿床学]