摘要
深度学习已被广泛应用于油气田开发领域的各个方面,但是纯数据驱动的深度学习模型存在数据需求量大、预测能力不稳定和泛化能力弱等问题,而且模型无法考虑数据背后蕴藏的物理规律。针对油藏压力动态预测问题,建立了油藏渗流物理和数据联合驱动的压力场预测深度神经网络模型,将非均质油藏渗流数学模型以正则化的形式加入到损失函数中,使得模型既能够服从数据训练的结果,又遵守渗流物理方程的约束。结果表明:联合驱动的深度神经网络模型可以实现压力场数据的高效学习和准确预测。对比纯数据驱动的深度神经网络模型,联合驱动的深度神经网络模型预测值与参考值的误差可降低93.1%,决定系数提高20.3%。在观测数据具有噪声的情况下,联合驱动的深度神经网络模型仍然可以保持较高的稳定性,具有较强的抗噪能力。
Deep learning has been widely used in various aspects of oil and gas field development.Nevertheless,purely data-driven deep learning models suffer from large amounts of required data,unstable prediction ability,and weak generalization ability,and such models fail to consider the physical laws underlying the data.For dynamic reservoir pressure prediction,a deep neural network model for pressure field prediction driven jointly by reservoir seepage physics and data was constructed,and the mathematical seepage model of heterogeneous reservoirs was added to the loss function through regularization,which enabled the model to conform to both the results of data training and the constraint of the seepage physics equations.The results show that the jointly driven deep neural network model can achieve efficient learning and accurate prediction of pressure field data.Compared with the purely data-driven deep neural network model,the jointly driven model reduces the error(L2)between predicted and reference values by 93.1%and increases the decision coefficient(R2)by 20.3%.In the case of noisy observed data,the jointly driven model can still maintain high stability with strong noise resistance.
作者
薛亮
戴城
韩江峡
杨明瑾
刘月田
XUE Liang;DAI Cheng;HAN Jiangxia;YANG Mingjin;LIU Yuetian(State Key Laboratory of Petroleum Resources and Prospecting,China University of Petroleum(Beijing),Beijing City,102249,China;College of Petroleum Engineering,China University of Petroleum(Beijing),Beijing City,102249,China;SINOPEC Petroleum Exploration and Production Research Institute,Beijing City,100083,China)
出处
《油气地质与采收率》
CAS
CSCD
北大核心
2022年第1期145-151,共7页
Petroleum Geology and Recovery Efficiency
基金
北京市自然科学基金面上项目“基于深度学习方法的致密气渗流高效随机模拟研究”(3222037)
中国石油科技创新基金项目“致密砂岩油藏低盐度水驱提高采收率机理研究”(2020D-5007-0203)
中国石油大学(北京)科研基金项目“基于大数据和机器学习的裂缝性油藏产能预测研究”(2462018QZDX13)
“具有油藏物理意识的深度神经网络研究”(2462021YXZZ010)
“微纳米孔隙油气流动微尺度效应”(2462020YXZZ028)。
关键词
深度神经网络
渗流物理
非均质性
正则化
损失函数
deep neural network
seepage physics
heterogeneity
regularization
loss function
