基于物理信息约束的拉普拉斯卷积神经网络渗流方程高效求解

朱海锋; 陈志明; 张宝允

doi:10.7623/syxb202609009

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于物理信息约束的拉普拉斯卷积神经网络渗流方程高效求解

油田开发 | 更新时间：2026-10-17

- 基于物理信息约束的拉普拉斯卷积神经网络渗流方程高效求解
- Efficient solution of flow equations using a physics-informed Laplace convolutional neural network
- 石油学报 2026年47卷第9期页码：1862-1873
- 作者机构：
  
  1. 中国石油大学（北京）石油工程学院,北京,102249
  2. 中国石油大学（北京）人工智能学院,北京,102249
- 作者简介：
- 基金信息：
- DOI：10.7623/syxb202609009
  中图分类号： TE319
- 网络首发：2026-10-14，
  
  纸质出版：2026
- 稿件说明：
移动端阅览
朱海锋, 陈志明, Wei Yu, 张宝允. 基于物理信息约束的拉普拉斯卷积神经网络渗流方程高效求解[J]. 石油学报, 2026, 47(9): 1862-1873.

Zhu Haifeng, Chen Zhiming, Wei Yu. Efficient solution of flow equations using a physics-informed Laplace convolutional neural network[J]. ACTA PETROLEI SINICA, 2026, 47(9): 1862-1873.
朱海锋, 陈志明, Wei Yu, 张宝允. 基于物理信息约束的拉普拉斯卷积神经网络渗流方程高效求解[J]. 石油学报, 2026, 47(9): 1862-1873. DOI： 10.7623/syxb202609009.

Zhu Haifeng, Chen Zhiming, Wei Yu. Efficient solution of flow equations using a physics-informed Laplace convolutional neural network[J]. ACTA PETROLEI SINICA, 2026, 47(9): 1862-1873. DOI： 10.7623/syxb202609009.

摘要

人工智能在油气开发领域的应用逐步深入，在油藏数值模拟方面也展现出良好前景。渗流偏微分方程是油藏数值模拟的核心工具，但由于其机理复杂且计算耗时，其高效准确求解仍是油藏数值模拟智能化的关键。近年来随着油气人工智能技术的发展，以卷积神经网络（CNN）为代表的深度学习模型，因其对空间分布数据表征能力强，为渗流方程求解提供了新范式。基于此，提出了一种基于物理信息约束的拉普拉斯卷积神经网络（PILCNN）的渗流方程求解方法。该方法通过在常规CNN模块中嵌入拉普拉斯卷积算子，实现对渗流场局部特征的精细捕捉，并利用有限体积法构建损失函数，约束神经网络训练，实现渗流方程的求解。针对油气藏常见的均质、非均质、混合断层边界及变制度生产4个场景开展验证。结果表明：PILCNN在均质及非均质问题上的预测平均相对误差低至0.009.7%和0.005.9%，与参考解高度一致；对于混合边界及变制度生产问题，PILCNN仍表现出良好的稳定性。进一步基于不同参数开展了4组参数（渗透率、边界距离、井储系数及表皮系数）的敏感性分析。研究结果显示，PILCNN在不同参数下的求解稳定性良好。此外，对比了不同物理信息约束的卷积神经网络、残差神经网络及全连接神经网络的训练效果，PILCNN在全训练周期表现更优。

Abstract

The application of artificial intelligence in oil and gas development has been progressively expanding and demonstrates considerable potential in reservoir numerical simulation. Flow partial differential equations (PDEs) serve as the core tool for reservoir numerical simulation. However

due to the complex computational mechanisms and high computational cost of PDEs

obtaining efficient and accurate solutions to PDEs is the key to intelligent reservoir numerical simulation. In recent years

driven by the rapid development of artificial intelligence technologies in the oil and gas industry

deep learning models

particularly convolutional neural networks (CNNs)

have emerged as a novel paradigm for solving flow equations owing to their powerful capability in capturing and representing spatially distributed data. Consequently

a Physical-Informed Laplace Convolutional Neural Network (PILCNN) is proposed for solving the flow equation. By integrating a Laplace convolution operator into the conventional CNN architecture

the proposed method enables the accurate extraction of local features within flow fields. Meanwhile

the loss function is formulated via the finite volume method (FVM) to enforce physical constraints during neural network training

thereby facilitating the solution of flow equations. The proposed method is validated under four common scenarios in oil and gas reservoirs

including homogeneous reservoirs

heterogeneous reservoirs

reservoirs with mixed fault boundaries

and variable-rate production conditions. The results indicate that PILCNN achieves mean relative errors (MRE) as low as 0.009.7% and 0.005.9% in predicting homogeneous and heterogeneous cases

respectively

showing excellent agreement with the reference solutions. In addition

PILCNN exhibits robust stability when applied to mixed boundary and variable-rate production problems. Furthermore

four groups of parameter sensitivity analyses are conducted by varying permeability

boundary distance

wellbore storage coefficient

and skin factor. The results show that PILCNN maintains reliable solution stability under different parameter configurations. Comparative training experiments are performed among the Physics-Informed Convolutional Neural Network (PICNN)

Physics-Informed Residual Neural Network (PIResNet) and Physics-Informed Neural Network (PIetNN). PILCNN demonstrates preferable performance throughout the entire training process.

关键词

Keywords

references

陈志明,YU Wei,石璐铭,等. 压裂水平井的多模式裂缝网络试井模型及参数评价——以吉木萨尔页岩油为例[J].石油学报,2023,44(10):1706-1726.

CHEN Xu,ZHANG Kai,JI Zhenning,et al.Progress and challenges of integrated machine learning and traditional numerical algorithms:taking reservoir numerical simulation as an example[J].Mathematics,2023,11(21):4418.

RAISSI M,PERDIKARIS P,KARNIADAKIS G E.Physics-informed neural networks:a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations[J].Journal of Computational Physics,2019,378:686-707.

徐宝昌,张学智,王雅欣,等.用于两相流环空压力预测的自适应物理信息神经网络模型[J].石油学报,2023,44(3):545-555.

刘雪琪,王志远,魏强,等.物理信息神经网络驱动的井筒温度场求解方法[J].石油学报,2025,46(2):413-425.

SUN Luning,GAO Han,PAN Shaowu,et al.Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data[J].Computer Methods in Applied Mechanics and Engineering,2020,361:112732.

HAO Weiqi,TAN Long,YANG Xiaoguang,et al.A physics-informed machine learning approach for notch fatigue evaluation of alloys used in aerospace[J].International Journal of Fatigue,2023,170:107536.

LAUBSCHER R.Simulation of multi-species flow and heat transfer using physics-informed neural networks[J].Physics of Fluids,2021,33(8):087101.

LIU Aodi,LI Jing,BI Jianfei,et al.A novel reservoir simulation model based on physics informed neural networks[J].Physics of Fluids,2024,36(11):116617.

BUCKLEY S E,LEVERETT M C.Mechanism of fluid displacement in sands[J].Transactions of the AIME,1942,146(1):107-116.

FUKS O,TCHELEPI H A.Limitations of physics informed machine learning for nonlinear two-phase transport in porous media[J].Journal of Machine Learning for Modeling and Computing,2020,1(1):19-37.

FRACES C G,PAPAIOANNOU A,TCHELEPI H.Physics informed deep learning for transport in porous media.Buckley Leverett problem[J].arXiv preprint arXiv:2001.05172,2020.

ALMAJID M M,ABU-AL-SAUD M O.Prediction of porous media fluid flow using physics informed neural networks[J].Journal of Petroleum Science and Engineering,2022,208:109205.

ZHANG Jingjing,BRAGA-NETO U,GILDIN E.Advancing petroleum engineering solutions:integrating physics-informed neural networks for enhanced Buckley-Leverett model analysis[J].arXiv preprint arXiv:2404.12502,2024.

ROCHA COUTINHO E J.Embedding reservoir physics into machine learning[D].College Station:Texas A&M University,2022.

HAN Jiangxia,XUE Liang,WEI Yunsheng,et al.Physics-informed neural network-based petroleum reservoir simulation with sparse data using domain decomposition[J].Petroleum Science,2023,20(6):3450-3460.

LI Daolun,SHEN Luhang,ZHA Wenshu,et al.Physics-constrained deep learning for solving seepage equation[J].Journal of Petroleum Science and Engineering,2021,206:109046.

LI Daolun,SHEN Luhang,ZHA Wenshu,et al.Deep learning-based method for solving seepage equation under unsteady boundary[J].International Journal for Numerical Methods in Fluids,2024,96(1):87-101.

SHEN Luhang,LI Daolun,ZHA Wenshu,et al.Physical Asymptotic-Solution nets:physics-driven neural networks solve seepage equations as traditional numerical solution behaves[J].Physics of Fluids,2023,35(2):023603.

BADAWI D,GILDIN E.Physics-informed neural network for the transient diffusivity equation in reservoir engineering[J].arXiv preprint arXiv:2309.17345,2023.

YUAN Biao,WANG He,HEITOR A,et al.f-PICNN:a physics-informed convolutional neural network for partial differential equations with space-time domain[J].Journal of Computational Physics,2024,515:113284.

ZHOU Wanyun,CHU Xiaowen.Auto-PICNN:automated machine learning for physics-informed convolutional neural networks[J].arXiv preprint arXiv:2407.06151,2024.

LI Daolun,WANG Huan,ZHA Wenshu,et al.Learning a CNN with the finite volume method for modelling fluid seepage flow under closed boundary[J].Geoenergy Science and Engineering,2024,243:213388.

LI Jianfeng,ZHOU Liangying,SUN Jingwei,et al.Physically plausible and conservative solutions to Navier-Stokes equations using physics-informed CNNs[J].JUSTC,2024,54(4):0403.

ZENG Lixiong,ZHANG Feng,LANG Genfeng,et al.Application of frequency aware mechanism-based physical information convolutional neural network in rolling bearing fault diagnosis[J].IEEE Sensors Journal,2025,25(10):16797-16811.

SUI Xiaodong,ZHANG Ru,LUO Yaozhi,et al.Multiple bolt looseness detection using SH-typed guided waves:integrating physical mechanism with monitoring data[J].Ultrasonics,2025,150:107601.

ZHANG Zhao.A physics-informed deep convolutional neural network for simulating and predicting transient Darcy flows in heterogeneous reservoirs without labeled data[J].Journal of Petroleum Science and Engineering,2022,211:110179.

ZHANG Zhao,YAN Xia,LIU Piyang,et al.A physics-informed convolutional neural network for the simulation and prediction of two-phase Darcy flows in heterogeneous porous media[J].Journal of Computational Physics,2023,477:111919.

CHEN Jungang,GILDIN E,KILLOUGH J E.Transfer learning-based physics-informed convolutional neural network for simulating flow in porous media with time-varying controls[J].Mathematics,2024,12(20):3281.

Lü Shuaijun,LI Daolun,ZHA Wenshu,et al.Solving seepage equation using physics-informed residual network without labeled data[J].Computer Methods in Applied Mechanics and Engineering,2024,418:116563.

HAN Jiangxia,XUE Liang,JIA Ying,et al.Prediction of porous media fluid flow with spatial heterogeneity using criss-cross physics-informed convolutional neural networks[J].CMES-Computer Modeling in Engineering and Sciences,2023,138(2):1323-1340.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

深床滤池内过滤与反洗过程流动规律研究

电解水制氧耦合绝氮燃烧与传统水泥生产技术的数值模拟对比

基于流场调控的PEM电解槽多目标性能优化

质子交换膜水电解制氢双极板流场研究进展

塔西北地区阿南区块、柯坪区块、乌什区块油气成藏模式与有利区优选