华南理工大学轻工科学与工程学院,广东 广州 510641
蔡至憧(2000—),男,硕士研究生,1551476983@qq.com
熊勤钢(1982—),男,博士,教授,qingangxiong@scut.edu.cn
收稿:2025-12-24,
修回:2026-01-28,
纸质出版:2026-10-25
移动端阅览
蔡至憧, 刘晓雯, 熊勤钢. 基于DNS和高量描述符ANN的多孔介质渗透率预测[J]. 化工学报, 2026, 77(10): 5787-5800
CAI Zhichong, LIU Xiaowen, XIONG Qingang. Permeability prediction of porous media based on DNS and high-dimensional descriptor ANN[J]. CIESC Journal, 2026, 77(10): 5787-5800
蔡至憧, 刘晓雯, 熊勤钢. 基于DNS和高量描述符ANN的多孔介质渗透率预测[J]. 化工学报, 2026, 77(10): 5787-5800 DOI: 10.11949/0438-1157.20251451.
CAI Zhichong, LIU Xiaowen, XIONG Qingang. Permeability prediction of porous media based on DNS and high-dimensional descriptor ANN[J]. CIESC Journal, 2026, 77(10): 5787-5800 DOI: 10.11949/0438-1157.20251451.
多孔介质渗透率的精确预测对能源开发、环境工程等领域具有重要意义。针对传统经验公式预测精确度不足的问题,提出了一种基于人工神经网络(artificial neural network,ANN)的渗透率预测模型。通过格子玻尔兹曼方法(lattice Boltzmann method,LBM)模拟620组人工多孔介质渗流过程获取其渗透率数据,结合MATLAB程序与分水岭分割(subnetwork of the oversegmented watershed,SNOW)算法量化表征孔隙结构,提取32项空间结构特征参数。采用L1正则化和递归特征消除法优化特征子集,发现总周长、孔隙率、平均喉道直径等11项关键参数对渗透率影响显著。通过对比,确定神经网络结构为11-55-44-33-22-1。交叉验证表明,模型预测决定系数(
R
²)均高于0.95,平均绝对百分比误差(mean absolute percentage error,MAPE)均小于4%。该模型在简单性与精确性之间实现了较好的平衡,且体现出良好的泛化能力,为多孔介质“结构-渗透率”关系提供了新见解,可以作为多孔介质渗透率预测的有力替代方案。
Accurate prediction of permeability in porous media is of great significance in fields such as energy development and environmental engineering. To address the problem of insufficient prediction accuracy of traditional empirical formulas
a permeability prediction model based on artificial neural network (ANN) is proposed. Permeability data of 620 artificial porous media samples were obtained by simulating their seepage processes using the lattice Boltzmann method (LBM). Combined with MATLAB programs and the subnetwork of the oversegmented watershed (SNOW) algorithm
the pore structures were quantitatively characterized and 32 spatial structural feature parameters were extracted. The feature subset is optimized using L1 regularization and recursive feature elimination
revealing that 11 key parameters
including total perimeter
porosity
and average throat diameter
significantly affect permeability. Through comparative analysis
the neural network architecture was determined to be 11-55-44-33-22-1. Cross-validation results demonstrated that the model's coefficient of determination (
R
²) consistently exceeded 0.95
with a mean absolute percentage error (MAPE) of less than 4%. This model achieves an excellent balance between simplicity and accuracy
exhibits strong generalization capability
provides new insights into the “structure-permeability” relationship of porous media
and can serve as a powerful alternative for permeability prediction in porous media.
Zhang X T , Ning G Y , Chen P , et al . Realizing mechanically robust and electrically conductive wood via vacuum pressure impregnation [J ] . ACS Applied Bio Materials , 2020 , 3 ( 9 ): 6071 - 6078 .
Wang Y D , Xue J K , Wang D S , et al . Lattice Boltzmann method for simulation of shale gas flow in kerogen nano-pores considering temperature dependent adsorption [J ] . International Journal of Oil, Gas and Coal Technology , 2020 , 23 ( 4 ): 409 - 426 .
Tian Z W , Wang J Y . Lattice Boltzmann simulation of CO 2 reactive transport in network fractured media [J ] . Water Resources Research , 2017 , 53 ( 8 ): 7366 - 7381 .
王灵权 . 格子Boltzmann方法在多孔介质流中的多尺度应用研究 [D ] . 重庆 : 重庆大学 , 2017 .
Wang L Q . A research on the lattice Boltzmann method for its application in multi-scale porous flows [D ] . Chongqing : Chongqing University , 2017 .
Kozeny J . Uber kapillare leitung des wassers im boden [J ] . Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften , 1927 , 136 ( 2 a): 271 - 306 .
Carman P C . Fluid flow through granular beds [J ] . Chemical Engineering Research and Design , 1997 , 75 : S32 - S48 .
Timur A . An investigation of permeability, porosity, & residual water saturation relationships for sandstone reservoirs [J ] . The Log Analyst , 1968 , 9 ( 4 ): 8 - 17 .
Coates G R , Xiao L Z , Prammer M G . NMR Logging: Principles and Applications [M ] . Houston : Halliburton Energy Services , 1999 : 45 - 67 .
Costa A . Permeability-porosity relationship: a reexamination of the Kozeny-Carman equation based on a fractal pore-space geometry assumption [J ] . Geophysical Research Letters , 2006 , 33 ( 2 ): 2005GL025134 .
Cui X , Bustin A M M , Bustin R M . Measurements of gas permeability and diffusivity of tight reservoir rocks: different approaches and their applications [J ] . Geofluids , 2009 , 9 ( 3 ): 208 - 223 .
Darabi H , Ettehad A , Javadpour F , et al . Gas flow in ultra-tight shale strata [J ] . Journal of Fluid Mechanics , 2012 , 710 : 641 - 658 .
Nabipour I , Mohammadzadeh-Shirazi M , Raoof A , et al . A data-driven approach for efficient prediction of permeability of porous rocks by combining multiscale imaging and machine learning [J ] . Transport in Porous Media , 2025 , 152 ( 4 ): 29 .
Kamrava S , Tahmasebi P , Sahimi M . Linking morphology of porous media to their macroscopic permeability by deep learning [J ] . Transport in Porous Media , 2020 , 131 ( 2 ): 427 - 448 .
Wang Y , Zhang S X , Ma Z , et al . Artificial neural network model development for prediction of nonlinear flow in porous media [J ] . Powder Technology , 2020 , 373 : 274 - 288 .
Yalamanchi P , Datta Gupta S . Estimation of pore structure and permeability in tight carbonate reservoir based on machine learning (ML) algorithm using SEM images of Jaisalmer sub-basin, India [J ] . Scientific Reports , 2024 , 14 : 930 .
Zhang J , Ma G , Yang Z B , et al . Knowledge extraction via machine learning guides a topology-based permeability prediction model [J ] . Water Resources Research , 2024 , 60 ( 7 ): e2024WR037124 .
Ma S J , Liu J F , Lin Y J , et al . Prediction of permeability of various geotechnical materials under different temperatures based on physical characteristics and machine learning [J ] . Fuel , 2025 , 379 : 133109 .
Kang Q , Li K Q , Fu J L , et al . Hybrid LBM and machine learning algorithms for permeability prediction of porous media: a comparative study [J ] . Computers and Geotechnics , 2024 , 168 : 106163 .
侯亚祺 , 张玮 , 张鸿 , 等 . 基于机器学习与粒子群算法的LBM多相流模型优化 [J ] . 化工学报 , 2025 , 76 ( 3 ): 1120 - 1132 .
Hou Y Q , Zhang W , Zhang H , et al . Optimization of LBM multiphase flow models based on machine learning and particle swarm algorithm [J ] . CIESC Journal , 2025 , 76 ( 3 ): 1120 - 1132 .
Gostick J T . Versatile and efficient pore network extraction method using marker-based watershed segmentation [J ] . Physical Review E , 2017 , 96 ( 2 ): 023307 .
Wang M R , Wang J K , Pan N , et al . Mesoscopic predictions of the effective thermal conductivity for microscale random porous media [J ] . Physical Review E , 2007 , 75 ( 3 ): 036702 .
何雅玲 , 王勇 , 李庆 . 格子Boltzmann方法的理论及应用 [M ] . 北京 : 科学出版社 , 2009 : 31 - 51 .
He Y L , Wang Y , Li Q . Lattice Boltzmann Method: Theory and Applications [M ] . Beijing : Science Press , 2009 : 31 - 51 .
Yin P J , Song H H , Ma H R , et al . The modification of the Kozeny-Carman equation through the lattice Boltzmann simulation and experimental verification [J ] . Journal of Hydrology , 2022 , 609 : 127738 .
阙云 , 陈宪 , 蒋国平 , 等 . 基于LBM的2D/3D原状花岗岩残积土细观渗流模拟 [J ] . 应用基础与工程科学学报 , 2024 , 32 ( 1 ): 193 - 207 .
Que Y , Chen X , Jiang G P , et al . Simulation of 2D/3D mesoscopic seepage in undisturbed granite residual soils based on the LBM [J ] . Journal of Basic Science and Engineering , 2024 , 32 ( 1 ): 193 - 207 .
Qian Y H , D'Humières D , Lallemand P . Lattice BGK models for Navier-Stokes equation [J ] . Europhysics Letters (EPL) , 1992 , 17 ( 6 ): 479 - 484 .
刘洋 . 基于格子Boltzmann方法的多孔介质微观渗吸机理研究 [D ] . 武汉 : 中国地质大学 , 2022 .
Liu Y . Microscopic mechanism study of imbibition in porous media based on lattice Boltzmann method [D ] . Wuhan : China University of Geosciences , 2022 .
郭照立 , 郑楚光 . 格子Boltzmann方法的原理及应用 [M ] . 北京 : 科学出版社 , 2009 : 57 - 74 .
Guo Z L , Zheng C G . Theory and Applications of Lattice Boltzmann Method [M ] . Beijing : Science Press , 2009 : 57 - 74 .
Ghia U , Ghia K N , Shin C T . High- Re solutions for incompressible flow using the Navier-Stokes equations and a multigrid method [J ] . Journal of Computational Physics , 1982 , 48 ( 3 ): 387 - 411 .
任俊杰 . 基于格子Boltzmann方法的页岩气微观流动机理研究 [D ] . 成都 : 西南石油大学 , 2015 .
Ren J J . Study of microscopic transport mechanisms of shale gas using the lattice Boltzmann method [D ] . Chengdu : Southwest Petroleum University , 2015 .
林涛 . 多孔介质内单相流动与传热的格子Boltzmann数值研究 [D ] . 东营 : 中国石油大学(华东) , 2016 .
Lin T . Numerical study on single phase flow and heat transfer in porous media with lattice Boltzmann method [D ] . Dongying : China University of Petroleum , 2016 .
谢颖颖 . 利用格子Boltzmann方法预测多孔介质中非牛顿流体的渗透率 [D ] . 武汉 : 华中科技大学 , 2020 .
Xie Y Y . A study on the lattice Boltzmann models for predicting effective permeability of non-Newtonian fluid in porous media [D ] . Wuhan : Huazhong University of Science and Technology , 2020 .
Hou S L , Zou Q S , Chen S Y , et al . Simulation of cavity flow by the lattice Boltzmann method [J ] . Journal of Computational Physics , 1995 , 118 ( 2 ): 329 - 347 .
Basheer I A , Hajmeer M . Artificial neural networks: fundamentals, computing, design, and application [J ] . Journal of Microbiological Methods , 2000 , 43 ( 1 ): 3 - 31 .
Nair V , Hinton G E . Rectified linear units improve restricted Boltzmann machines [C ] // Proceedings of the 27th International Conference on International Conference on Machine Learning . Haifa, Israel : ACM , 2010 : 807 - 814 .
Kingma D P , Ba J . Adam: a method for stochastic optimization [PP/OL ] . V9.arxiv ( 2017-01-30 )[ 2026-01-28 ] . https://doi.org/10.48550/arXiv.1412.6980 https://doi.org/10.48550/arXiv.1412.6980 .
Tibshirani R . Regression shrinkage and selection via the lasso: a retrospective [J ] . Journal of the Royal Statistical Society Series B: Statistical Methodology , 2011 , 73 ( 3 ): 273 - 282 .
郁伯铭 . 分形多孔介质输运物理 [M ] . 北京 : 科学出版社 , 2014 : 8 - 44 .
Yu B M . Transport Physics of Fractal Porous Media [M ] . Beijing : Science Press , 2014 : 8 - 44 .
van der Linden J H , Narsilio G A , Tordesillas A . Machine learning framework for analysis of transport through complex networks in porous, granular media: a focus on permeability [J ] . Physical Review E , 2016 , 94 ( 2 ): 022904 .
孔祥言 . 高等渗流力学 [M ] . 2版 . 合肥 : 中国科学技术大学出版社 , 2010 : 26 - 53 .
Kong X Y . Advanced Seepage Mechanics [M ] . 2nd ed . Hefei : University of Science and Technology of China Press , 2010 : 26 - 53 .
Yasuda T , Ookawara S , Yoshikawa S , et al . Machine learning and data-driven characterization framework for porous materials: permeability prediction and channeling defect detection [J ] . Chemical Engineering Journal , 2021 , 420 : 130069 .
朱维耀 , 王玉普 , 邓庆军 . 多孔介质细观流动理论及模拟方法 [M ] . 北京 : 科学出版社 , 2021 : 1 - 25 .
Zhu W Y , Wang Y P , Deng Q J . Mesoscopic Flow Theory and Simulation Method in Porous Media [M ] . Beijing : Science Press , 2021 : 1 - 25 .
Nie F , Wang H C , Zhao Y X , et al . A universal correlation for flow condensation heat transfer in horizontal tubes based on machine learning [J ] . International Journal of Thermal Sciences , 2023 , 184 : 107994 .
Nguyen Q , Mondelli M . Global convergence of deep networks with one wide layer followed by pyramidal topology [C ] // Proceedings of the 34th International Conference on Neural Information Processing Systems , 2020 : 11961 - 11972 .
Zhang G Y , Wang Z Z , Mohaghegh S , et al . Pattern visualization and understanding of machine learning models for permeability prediction in tight sandstone reservoirs [J ] . Journal of Petroleum Science and Engineering , 2021 , 200 : 108142 .
Kumar I , Tripathi B K , Singh A . Velocity field prediction for digital porous media using attention-based deep convolutional recurrent neural network [J ] . Chemical Engineering Science , 2025 , 314 : 121825 .
Takbiri S , Kazemi M , Takbiri-Borujeni A , et al . A deep learning approach to predicting permeability of porous media [J ] . Journal of
0
浏览量
10
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010102001995号