分频属性反演方法在B油田的应用  

Application of frequency division attribute inversion method in B oilfield

作　　者：王奇 WANG Qi(Geology Research Institute,CNPC Great Wall Drilling Co.,Ltd.,Panjin 124010,Liaoning,China)

机构地区：[1]中国石油集团长城钻探工程有限公司地质研究院,辽宁盘锦124010

出　　处：《矿产与地质》2020年第3期567-570,共4页Mineral Resources and Geology

摘　　要：针对B油田常规声波阻抗曲线难以识别砂泥岩的储层预测难题,通过多曲线交汇、敏感性分析,确定中子-密度属性反演预测储层的基本思路。在建立三维低频地质模型基础上,对地震资料进行分频处理,利用神经网络方法计算不同厚度下振幅与频率之间的关系,建立测井曲线与地震波形之间的非线性关系,最终摸索出基于精细地质模型的神经网络分频属性反演方法。研究结果表明:该方法解决了复杂储层预测的难题,准确预测了该油田储层的变化规律,为下步开发调整方案编制提供了重要依据。According to multiple curve intersection and sensitivity analysis,the basic idea of neutron-density attribute inversion is determined in order to solve the problem that conventional acoustic impedance curve is difficult to identify sand-shale reservoir formation in B oilfield.Based on the establishment of a 3D low-frequency geological model,the seismic data are processed by frequency division.The relationship between amplitude and frequency under different thickness is calculated by neural network method,the nonlinear relationship between logging curve and seismic waveform is established,and finally the neural network frequency division attribute inversion method based on the fine geological model is found out.The result shows that this method solves the difficult problem of complex reservoir formation prediction,accurately predicts the variation regularity of the reservoir formation,and provides an important basis for the next development adjustment plan.

关 键 词：分频属性反演 地质模型 中子密度交汇 神经网络 

分 类 号：P618.13[天文地球—矿床学] P631.4[天文地球—地质学]