油气藏裂缝型储层预测方法——以济阳坳陷古潜山为例

济阳坳陷富台油田的潜山主要由下古生界寒武系和奥陶系碳酸盐岩地层组成。采用特征重构和属性反演技术，分析了碳酸盐岩储层发育机理，认为其储层裂缝的测井响应特征主要表现为深侧向电阻率（R_lld）变化不大，浅侧向电阻率（R_lls） 的值相对降低。通过重构反映储层裂缝发育程度的特征曲线，并选取相关的地震属性作为井间信息，用概率神经网络反演获得反映裂缝发育程度的特征属性数据体，预测裂缝发育带的空间展布。预测结果表明，所述方法具有良好效果。

PREDICTION METHODS OF FRACTURED RESERVOIR: TAKING FOSSIL BURIED HILL IN JIYANG DEPRESSION AS AN EXAMPLE

Buried hills of Futai oilfield in Jiyang depression are mainly composed of Cambrian and Ordovician carbonates. In order to accurately predict the spatial distribution of fractured zone in the buried hill reservoir in the study area, feature reconfiguration and attribute inversion techniques are used to analyze the mechanism of carbonate development. The log responses of fractures are mainly characterized by small change of deep laterolog resistivity and relative decline of shallow laterolog resistivity. The characteristic curve reflecting fracture development is reconfigured and the spatial distribution of fractured zone is predicted by taking relevant seismic attributes as interwell information and inverting characteristic attribute data volume reflecting fracture development with probability neural network. It is predicted that the fracture porosity of Yeli-Liangjia-Fengshan Fm is relatively high in the structural axis and near C204 well to the south. This prediction result is verified by C204-5 well drilled later, which penetrates well-developed fractured reservoirs and produces 47.3 t oil equivalent per day.