基于岩石物理的致密砂岩油藏叠前优势储层预测——以渤海湾盆地南堡凹陷4号构造东营组二段为例

针对致密砂岩油层与围岩波阻抗叠置严重、油层与围岩区分难度大等问题,应用岩石物理建模与分析、地震正演模拟、叠前弹性参数反演、贝叶斯岩相判别等技术对渤海湾盆地南堡凹陷4号构造东营组二段致密砂岩油藏优势储层进行预测。在测井资料预处理基础上,依据Biot-Gassmann理论,对Xu-White模型所需岩石物理参数进行计算,并正演估算弹性参数。采用Biot-Gassmann方程与Brie经验公式进行流体替换,建立岩石物理解释量版,进而优选出致密砂岩储层敏感弹性参数;同时根据不同地质模型与AVO正演模拟,确定储层预测极限厚度、有效目标和反演方法。最后利用贝叶斯判别方法计算叠前弹性参数反演成果的岩相类型概率分布,对有利孔隙砂岩分布进行预测。结果表明:油藏状态下纵横波速度比和纵波阻抗为最敏感的岩石弹性参数,对砂岩、泥岩具有区分能力;储层预测的有效目标是地震可识别层段内具有厚度比例优势的有利孔隙砂岩,叠前弹性参数反演和岩性与流体概率分析是解决研究区优势储层预测的有效方法;应用该方法预测的有利孔隙砂岩分布与实钻井吻合度高,储层分布规律与研究区地质规律一致,提高了油层识别能力及预测精度。

Pre-stack favorable reservoir prediction in tight sandstone reservoirs based on rock physics:a case study of the 2nd Member of Dongying Formation,Structure 4,Nanpu sag,Bohai Bay Basin

Aiming at the serious wave impedance superposition between tight sandstone oil reservoirs and surrounding rocks, large difficulty in distinguishing oil-bearing layers and surrounding rocks and other problems, the technologies such as modeling and analysis, seismic forward modeling, pre-stack elastic parameter inversion, and Bayesian discrimination are used to predict the favorable reservoir of tight sandstone oil reservoir in the 2nd Member of Dongying Formation, Structure 4, Nanpu sag, Bohai Bay Basin. Based on logging data preprocessing and Biot-Gassmann theory, a calculation was performed on the petrophysical parameters required by Xu-White model, followed by a forward estimation on elastic parameters. Fluid substitution is conducted by Biot-Gassmann equation and Brie empirical equation, and then rock physics interpretation template is established to select the optimal sensitive elastic parameters of tight sandstone oil reservoir. Meanwhile, the limited thickness, effective target and inversion methods are determined for reservoir prediction according to the forward modeling between different geological models and AVO. Finally, Bayesian discriminating method is used to calculate the lithofacies types probability distribution of pre-stack elastic parameter inversion results, so as to predict the favorable porous sandstone distribution. The results demonstrate that under reservoir oil-bearing conditions, the ratio of P-wave velocity and S-wave velocity and P-wave impedance are the most sensitive rock elastic parameters to distinguish sandstone and mudstone. The effective target of reservoir prediction is the favorable porous sandstone with thickness proportional odds in the objective intervals which could be identified by seismic data. Pre-stack elastic parameter inversion and litho-fluid probability analysis are the effective methods for predicting favorable reservoirs in the study area. The favorable porous sandstone distribution predicted using the method in this study is highly consistent with the actual drilling results, and the reservoir distribution laws are in accordance with regional geological laws, greatly enhancing the oil-layer identification ability and prediction accuracy.