火成岩屏蔽下碎屑岩储层预测方法研究

Research of Prediction Methods for the Clastic Reservoir Under Igneous Rock Shielding

[目的]火成岩层对地震波有屏蔽和吸收作用,使得火成岩下伏地层的地震资料品质较差,储层边界难以识别。[方法]通过正演模拟不同火成岩在速度、厚度、与砂体距离等方面对下伏地层的影响,定量分析影响砂体尖灭点识别精度的因素,寻找恢复真实储层边界的描述方法。利用分频滤波去噪、弱信号恢复等解释性处理技术,压制火成岩干扰,提高弱反射储层的识别能力。在解释性处理的基础上,再综合测井、地质等多信息,建立井控多属性融合储层定量预测技术。[结果]综合利用以上技术成果,针对玉皇庙地区夏41-商97构造带,开展储层精细刻画,并取得了较好的效果。[结论]利用处理后的地震资料,对研究工区沙三上3砂组边界进行重新识别描述。处理后的剖面,砂体反射特征连续性更好,砂体边界继续向东延伸,较原始砂体尖灭线向东前移25道。将砂体描述结果与实钻井数据进行对比,描述吻合率达到85%。

[Purposes]The strong shielding and absorption of igneous rock to seismic wave result in poor quality of seismic data of the underlying strata.The seismic reflection of surrounding rock is changeable,so it is difficult to identify the reservoir boundary.[Methods]By forward modeling the influence of thickness and occurrence of igneous rocks on underlying strata,the factors affecting the accuracy of reservoir identification are quantitatively analyzed.On the basis of forward modeling model,interpretative processing techniques such as filtering,de-noising,well control and frequency raising,multi-wavelet decomposition and reconstruction are used to suppress igneous rock interference and to improve the identification ability of weak reflection reservoir.On the basis of interpretative processing,the well control multiattribute fusion reservoir quantitative prediction technology is established by integrating multiple logging geological information.[Findings]Based on the above technical achievements,fine reservoir characterization was carried out for the Xia41-Shang97 structural belt in Yuhuangmiao area,and good results were achieved.[Conclusions]Using the processed seismic data,the boundary of the upper 3rd sand group in the study area was re identified and described.After processing,the continuity of sand body reflection characteristics is better,and the boundary of the sand body continues to extend eastward,moving 25 channels ahead of the original sand body pinch out line.Comparing the sand body description results with actual drilling data,the description coincidence rate reaches 85%.