【摘要】 为了对钻井过程中的井壁失稳问题进行有效控制,根据地震和测井信息的分形性质以及它们之间的密切联系,提出了综合利用地震记录的分形属性预测测井曲线并进而在钻前预测安全钻井液密度范围的模型。首先从井旁地震记录中提取出能全面反映地震非线性特征的分形属性,运用能充分利用样本信息的CUSI神经网络,建立已钻井段地震分形属性与测井数据之间的非线性映射模型,通过该模型预测未钻地层的声波和密度测井曲线。在此基础上运用井壁稳定分析模型,计算地层的孔隙压力、坍塌压力和破裂压力,从而在钻前预测安全钻井液密度范围。该方法在西部某油田的应用表明,用地震记录的分形属性钻前预测安全钻井液密度范围的方法具有良好的预测效果。该方法预测精度较高,操作简单易行。更多还原

【Abstract】 The model comprehensively using seismic fractal attributes to predict logging curve and safe drilling fluid density range before drilling was presented based on the fractal properties and close relationship between them.The fractal attributes which can fully reflect seismic fractal characteristic were obtained from borehole-side seismic trace.Through CUSI neutral network which can completely utilize samples information,the nonlinear mapping model was constructed between seismic fractal attributes and logging data of drilled formation,which was then used to predict sonic and density logging under borehole bottom.On the basis of the model the analysis model of borehole stability were employed to calculate pore pressure,collapse pressure and fracture pressure,thus predicting safe drilling fluid density range before drilling so as to effectively control borehole instability.The practical application in oilfield indict that the prediction method was of high accuracy.更多还原