基于Mogi-Coulomb破坏准则的钻井安全压力窗口预测方法

Prediction Method of Drilling Safety Pressure Window Based on Mogi-Coulomb Failure Criteria

受拉伸和剪切破坏所影响的井眼失稳会导致严重的钻井问题,并增加油田的非生产时间和油气行业中的修井成本。为此,创新采用了依据Mogi-Coulomb破坏准则确定岩石强度的方法来预测所需的最佳钻井液密度,即钻井安全压力窗口。预测方法主要是通过Mogi-Coulomb模型中优化后的方程来计算钻井液密度窗口上下边界的坍塌压力和破裂压力。方法研究中的相关输入参数均取自油田高压地层,参数包括孔隙压力、原地应力数据和岩石力学特性。研究表明造成井眼失稳的主要原因是应力、岩石强度或孔隙压力的变化,且根据岩石破坏模型计算坍塌压力和破裂梯度的数值方法能够高效率、低成本地优选钻井液密度。本研究的目标是准确建立钻井安全压力操作窗口,使设计的钻井液密度既能实现有效钻进,又能维持井壁稳定性,防止钻井液漏失。在验证应用部分中通过输入墨西哥湾地区实验井的特性参数,比较预测和实际钻井液密度随深度的变化关系图,证明了所建立的模型具有较高准确度。

Borehole instability due to tensile and shear failure can lead to serious drilling problems and increase the non-production time of oilfield and the cost of workover in the oil and gas industry.In this pa-per,we innovatively adopt the method of determining the rock strength based on Mogi-Coulomb failure criterion to predict the best mud density,which is the drilling safety pressure window.For the prediction methods,optimized equation in Mogi-Coulomb model is mainly used to calculate the collapse pressure and fracture pressure at upper and lower boundary of the mud density window.The relevant input parameters in this study are all from high-pressure formations in the oilfield.The parameters include pore pressure,original in-situ stress data and rock mechanical properties.The results show that the main causes of borehole instability are the changes of stress,rock strength or pore pressure,and the numerical method of calculating the collapse pressure and fracture gradient based on the rock failure model is a faster and cheaper one to optimize the mud density.The objective of this study is to establish a drilling safety pressure operation window so that the designed mud density is high enough to get borehole stability and prevent mud loss.During the verification of the application,the relationship between the predicted and actual mud density and the depth is compared by inputting the characteristic parameters of the experimental wells in Maxico Gulf area,which proves the high accuracy of the model.