基于地震属性分析的地层孔隙压力钻前预测模型

钻前准确预测地层孔隙压力是进行井壁稳定状态分析的基础。基于地震和测井信息间的密切联系，提出了利用地震属性预测孔隙压力的新方法。通过神经网络建立已钻井段的井旁地震属性和声波时差测井数据之间的非线性关系模型，以此模型为基础预测未钻地层的声波速度。利用预测结果结合岩石力学模型得到垂直有效应力，最后根据有效应力原理计算地层孔隙压力。该方法有效克服了传统的地震层速度预测模型的不足之处，在油田的实际应用中取得了良好的效果，尤其适用于在勘探新区进行孔隙压力预测。

Model Used for Pre-drilling Prediction of Formation Pore Pressure Based on Seismic Attribute

Predicting pore pressure accurately is the foundation for analyzing borehole stability. Based on the close relationship between seismic and logging information, a new method for predicting pore pressure is presented based on seismic attribute analysis. Through neural network technology, the nonlinear relationship model is established between borehole-side seismic attributes and sonic logging data of drilled formation so as to predict acoustic velocity under bottomhole. On the basis of the predicted results and rock mechanical model, the vertical effective pressure is obtained. Pore pressure of formation can be calculated by using the effective pressure principle. This method, which effectively overcomes the weak points of the conventional prediction model based on seismic interval velocity data, has excellent results in practical application in the oilfield. It is particularly suitable for the prediction of pore pressure in new prospecting areas.