基于压缩感知和反演理论的储层预测研究

2022年 61卷 第No. 6期
阅读:88
查看详情
Reservoir prediction based on compressed sensing and inversion theory
(中海石油(中国)有限公司天津分公司渤海石油研究院,天津300450)

渤海A油田明上段河道砂体发育,是渤海唯一动用明上段地质储量的区块,但该工区地震资料采集时间早,资料中低频成分严重缺失,工区现有地球物理资料中明上段储层地震响应弱,根据这些资料无法指导水平井的部署与实施。为了恢复明上段储层的地震响应特征,指导明上段储层开发,开展了基于压缩感知和稀疏反演的储层预测研究。在用压缩感知理论求得反射系数序列后,将其变换到频率域,然后使用重构算法将反射系数序列的低频成分补偿到地震数据中,实现地震数据的低频补偿。对低频补偿后的地震数据进行波阻抗反演,所得反演结果改善了明上段储层的地震响应,反演剖面上储层顶、底界面相较于现有资料变得更易识别,与已钻井结果对比验证了研究成果的正确性。利用该研究成果,指导了渤海A油田明上段储层的井位部署,在油田开发中获得了较好的应用效果。

Bohai A Oilfield has well-developed channel sand bodies in the N2m formation,making it the only block in the Bohai Bay where geological reserves in the N2m formation are deposited.However,seismic data acquisition has not been performed recently in this area,and the low-frequency components of the data are lacking.Additionally,existing geophysical data in the work area exhibited poor seismic response to the reservoirs in the N2m formation,which hampered the deployment and implementation of horizontal wells.To restore the seismic response characteristics of channel reservoirs in the N2m formation and guide the development of N2m reservoirs,a reservoir prediction study was carried out based on compressive sensing and acoustic impedance inversion.The reflection coefficient sequence is obtained using the compressed sensing theory and then transformed into the frequency domain.Then,low-frequency compensation of the seismic data was carried out using the reconstruction algorithm.Acoustic impedance inversion was performed on the seismic data after low-frequency compensation,and the results improved the seismic response of channel reservoirs in the N2m formation.Compared with existing data,the top and bottom interfaces of the channel reservoirs became easier to identify.The accuracy of the results was verified via comparison with the drilled wells.The results of this paper were used to guide the well location deployment of N2m formation channel reservoirs in the Bohai A Oilfield and attained good application effects in oil field development.

低频成分缺失; 压缩感知; 地震资料低频补偿; 稀疏反演; 河道储层预测; 井位部署;
low frequency components loss;; compressed sensing;; low frequency compensation of seismic data;; sparse inversion;; channel reservoir prediction;; well deployment;

中海石油(中国)有限公司综合科研项目“渤海双高油田挖潜关键技术研究”(YXKY-2018-TJ-04)和中国海洋石油集团有限公司海油众创项目“利用振幅包络补偿地震低频信息”(HYZC-2022-01-093-TJ)共同资助。

10.3969/j.issn.1000-1441.2022.06.008