【摘要】 针对低渗透多孔介质中存在启动压力梯度的问题,分析了影响启动压力梯度的主要因素.采用BP人工神经网络的方法对启动压力梯度进行预测,并结合岩心实际测定的启动压力梯度进行验证.研究结果表明,BP人工神经网络是一种较为有效的预测方法,具有较高的精度.该方法的应用,可以为低渗油田的开发提供可靠的基础数据,节省人力物力.更多还原

【Abstract】 There is threshold pressure gradient (TPG) in low-permeability porous media. The main factors of influencing TPG are analyzed. It is put forward that TPG is predicted by BP neural network, and the predicted results are verified by practically measured data. The result shows that this prediction approach is very effective and has higher accuracy. The application of this approach can supply basic data for the development of low-permeability oilfields so as to save cost and labor.更多还原