【摘要】 勘探结果表明,兴隆台-马圈子地区沙三中下亚段发育浊积岩储层,其分布对油气具有明显控制作用,且具有较强隐蔽性,对有利储层展布规律认识不清制约了该区勘探部署工作。为此,首先综合分析录井岩性信息与SP、GR等敏感测井曲线,构建出能够较好反映岩性变化的岩性指示曲线;然后利用叠前AVO处理得到截距(P)、梯度(G)、截距*梯度(P*G)、远角低频(FLF)、密度(ρ)与剪切模量(μ)和拉梅常数(λ)的组合μρ、λρ等属性数据体,并提取了振幅、频率、相位等多种属性体;最后以岩性指示曲线为目标曲线,优选敏感性较好的一组叠前地震属性作为反演参数,利用神经网络反演技术对兴隆台-马圈子地区浊积岩储层进行了识别和描述,预测结果与已钻井及地质认识符合较好。更多还原

【Abstract】 Exploration results indicated that the turbidite reservoirs were developed in the mid-lower member of Es3 in Xinglongtai-Majuanzi Region,where oil and gas was apparently controlled by their distribution,and had a strong hidden nature.The arrangement of exploration was seriously restricted because of unclear recognition of favorable reservoir distribution.Therefore,the logging lithologic information and sensitive curves of SP,GR were comprehensively analyzed first and the lithology-indicated curve was established for reflecting the lithologic variations.Then,the pre-stack AVO process was carried out to obtain multiple attribute volumes,including intercept(P),gradient(G),PG,far angle low frequency(FLF)and μρ,λρ to extract seismic attributes including amplitude,frequency and phase.Finally,by taking the lithology-indicated curve as the target curve,the pre-stack seismic attributes with high sensitivity were optimized as inversion parameters,neural network inversion was used to identify and describe the turbidite reservoirs in Xinglongtai-Majuanzi Region.The result is consistent with that of drilled wells and geological recognitions.更多还原