

浏览全部资源
扫码关注微信
1.重庆交通大学河海学院,重庆 400074
2.重庆交通大学国家内河航道整治工程技术研究中心,重庆 400074
3.重庆交通大学水利水运工程教育部重点实验室,重庆 400074
4.重庆市综合交通运输研究所有限公司,重庆 401121
Received:17 April 2025,
Revised:2025-06-14,
Published:28 October 2025
移动端阅览
梁越,舒云林,刘港庆等.基于DL‑ERT模型的地下水渗透系数预测方法研究[J].防灾减灾工程学报,2025,45(05):1032-1041.
LIANG Yue,SHU Yunlin,LIU Gangqing,et al.Research on Prediction Method for Groundwater Permeability Coefficient Based on DL‑ERT Model[J].Journal of Disaster Prevention and Mitigation Engineering,2025,45(05):1032-1041.
梁越,舒云林,刘港庆等.基于DL‑ERT模型的地下水渗透系数预测方法研究[J].防灾减灾工程学报,2025,45(05):1032-1041. DOI: 10.13409/j.cnki.jdpme.20250417002.
LIANG Yue,SHU Yunlin,LIU Gangqing,et al.Research on Prediction Method for Groundwater Permeability Coefficient Based on DL‑ERT Model[J].Journal of Disaster Prevention and Mitigation Engineering,2025,45(05):1032-1041. DOI: 10.13409/j.cnki.jdpme.20250417002.
针对传统方法刻画地下水含水层非均质性时面临的精度不足、预测成本高等问题,基于数值模拟和室内砂箱试验,通过残差网络优化集成卷积门控循环单元(CNN‑GRU)的强大数据学习能力和电阻率层析成像法(ERT)运用物理先验信息的优势,提出一种融合物理机理的深度学习算法—DL‑ERT模型。将其对比多个传统反演模型,探讨融合算法在地下水含水层的渗透系数刻画精度。结果表明:(1)模型的训练损失和验证损失都快速下降并趋近于零,且两者的收敛几乎同步,表明DL‑ERT模型的构建策略优秀,能快速有效的学习数据特征;(2)以某一测试集样本为例,对比ERT、CNN‑GRU和DL‑ERT对该样本的渗透系数反演云图,发现单一的算法模型均不能同时注重左右两侧的高渗区域刻画,而DL‑ERT则对高渗区域表现出极大的预测潜力,其拟合精度达到了0.906;(3)制作室内砂箱试验,将融合算法与传统的克里金插值法、CNN‑GRU以及ERT作对比运用,得到各模型的拟合精度值分别为0.895、0.707、0.760和0.836。可以发现,DL‑ERT确实在一定程度上弥补了单一算法的不足,相比于单一的CNN‑GRU和ERT,其结果的预测精度提升了7%~17%,表明了该模型在工程运用方面的潜力。
To address the issues of insufficient accuracy and high prediction costs faced by conventional methods in characterizing the heterogeneity of groundwater aquifers
this study proposed a physics-informed deep learning algorithm—the DL-ERT model—based on numerical simulations and laboratory sandbox experiments. The model integrated the powerful data learning capability of a convolutional gated recurrent unit (CNN-GRU) optimized by residual networks with the advantage of physical prior information from electrical resistivity tomography (ERT). The DL-ERT model was compared with multiple traditional inversion models to examine the accuracy of the fusion algorithm in characterizing the permeability coefficient of groundwater aquifers. The results showed that: (1) the training and validation losses of the DL-ERT model rapidly decreased and approached zero
and their convergence was almost synchronous
indicating that the construction strategy of the DL-ERT model was excellent and that data features could be quickly and effectively learned. (2) Taking a sample from the test set as an example
the inversion cloud maps of the permeability coefficient obtained by ERT
CNN-GRU
and DL-ERT were compared. It was found that individual algorithm models could not simultaneously capture the high-permeability zones on both sides
while DL-ERT demonstrated remarkable predictive potential for high-permeability zones
achieving a fitting accuracy of 0.906. (3) Laboratory sandbox experiments were conducted
and the fusion algorithm was compared with traditional Kriging interpolation
CNN-GRU
and ERT
yielding fitting accuracies of 0.895
0.707
0.760
and 0.836
respectively. It is evident that the DL-ERT model compensates for the limitations of individual algorithms to some extent
with prediction accuracy improved by 7%-17% compared with the individual CNN-GRU and ERT models
indicating the potential of the model for engineering applications.
王鑫永 , 王锡勇 , 章杨松 , 等 . 岩体裂隙连通路径及岩体等效渗透张量参数研究 [J]. 防灾减灾工程学报 , 2020 , 40 ( 5 ): 789 ‑ 795 .
Wang X Y , Wang X Y , Zhang Y S , et al . Study on connected path of rock fracture and equivalent permeability tensor parameters of rock mass [J]. Journal of Disaster Prevention and Mitigation Engineering , 2020 , 40 ( 5 ): 789 ‑ 795 . (in Chinese)
文志杰 , 姜鹏飞 , 景所林 , 等 . 煤矿地下水库底板渗流模拟试验系统研制及验证 [J]. 煤炭学报 , 2021 , 46 ( 5 ): 1487 ‑ 1497 .
Wen Z J , Jiang P F , Jing S L , et al . Development and verification of simulation testing system for floor seepage in coal mine underground reservoir [J]. Journal of China Coal Society , 2021 , 46 ( 5 ): 1487 ‑ 1497 . (in Chinese)
江文豪 , 冯晨 , 黄啸 , 等 . 缓冲层-复合隔离墙-含水层系统中有机污染物二维运移规律研究 [J]. 岩土力学 , 2024 , 45 ( 增1 ): 731 ‑ 741 .
Jiang W H , Feng C , Huang X , et al . Two‑dimensional transport mechanisms of organic contaminants in a buffer layer‑composite cutoff wall‑aquifer system [J]. Rock and Soil Mechanics , 2024 , 45 ( Sup1 ): 731 ‑ 741 . (in Chinese)
曹渊 , 牛冠毅 , 王铁良 . 基于钻孔充气试验的岩石渗透率原位测量研究 [J]. 岩土工程学报 , 2017 , 39 ( 3 ): 534 ‑ 539 .
Cao Y , Niu G Y , Wang T L . In‑situ measurement of rock permeability based on pneumatic tests in boreholes [J]. Chinese Journal of Geotechnical Engineering , 2017 , 39 ( 3 ): 534 ‑ 539 . (in Chinese)
吴冰 , 朱鸿鹄 , 曹鼎峰 , 等 . 黄土水分场光纤原位监测及非饱和渗透系数估算 [J]. 防灾减灾工程学报 , 2019 , 39 ( 5 ): 691 ‑ 699 .
Wu B , Zhu H H , Cao D F , et al . In‑situ monitoring of moisture field and estimation of unsaturated permeability coefficient of loess foundation [J]. Journal of Disaster Prevention and Mitigation Engineering , 2019 , 39 ( 5 ): 691 ‑ 699 . (in Chinese)
包孟碟 , 朱俊高 , 吴二鲁 , 等 . 基于级配方程的粗粒土渗透系数经验公式及其验证 [J]. 岩土工程学报 , 2020 , 42 ( 8 ): 1571 ‑ 1576 .
Bao M D , Zhu J G , Wu E L , et al . Empirical formula for permeability coefficient of coarse grained soil based on gradation equation and its verification [J]. Chinese Journal of Geotechnical Engineering , 2020 , 42 ( 8 ): 1571 ‑ 1576 . (in Chinese)
蒋小虎 , 黄跃廷 , 胡海军 , 等 . 基于原位双环、试坑浸水试验和数值模拟反演的Q3黄土饱和渗透系数对比研究 [J]. 岩土力学 , 2022 , 43 ( 11 ): 2941 ‑ 2951 .
Jiang X H , Huang Y T , Hu H J , et al . Comparison of saturated permeability coefficient of Q3 loess based on in‑situ double ring test, field water immersion test and numerical simulation inversion [J]. Rock and Soil Mechanics , 2022 , 43 ( 11 ): 2941 ‑ 2951 . (in Chinese)
Eggleston J , Rojstaczer S . The value of grain-size hydraulic conductivity estimates: Comparison with high resolution in-situ field hydraulic conductivity [J]. Geophysical Research Letters , 2001 , 28 ( 22 ): 4255 ‑ 4258 .
Ruan K , Fu X L . A modified Kozeny-Carman equation for predicting saturated hydraulic conductivity of compacted bentonite in confined condition [J]. Journal of Rock Mechanics and Geotechnical Engineering , 2022 , 14 ( 3 ): 984 ‑ 993 .
Vienken T , Dietrich P . Field evaluation of methods for determining hydraulic conductivity from grain size data [J]. Journal of Hydrology , 2011 , 400 ( 1‑2 ): 58 ‑ 71 .
李治甫 , 康帅 , 王自法 , 等 . 基于时频变换和卷积神经网络的结构损伤识别 [J]. 防灾减灾工程学报 , 2023 , 43 ( 6 ): 1275 ‑ 1283 .
Li Z F , Kang S , Wang Z F , et al . Structural damage identification based on time‑frequency transform and convolutional neural network [J]. Journal of Disaster Prevention and Mitigation Engineering , 2023 , 43 ( 6 ): 1275 ‑ 1283 . (in Chinese)
张成龙 , 丁世飞 , 郭丽丽 , 等 . 随机配置网络研究进展 [J]. 软件学报 , 2024 , 35 ( 5 ): 2379 ‑ 2399 .
Zhang C L , Ding S F , Guo L L , et al . Research Progress on Stochastic Configuration Network [J]. Journal of Software , 2024 , 35 ( 5 ): 2379 ‑ 2399 . (in Chinese)
Srisutthiyakorn N . Deep‑learning methods for predicting permeability from 2D/3D binary‑segmented images [C]∥ SEG technical program expanded abstracts 2016 . Tulsa : Society of Exploration Geophysicists , 2016 : 3042 ‑ 3046 .
Zhong Z , Carr T R , Wu X , et al . Application of a convolutional neural network in permeability prediction: A case study in the Jacksonburg‑Stringtown oil field, West Virginia, USA [J]. Geophysics , 2019 , 84 ( 6 ): B363 ‑ B373 .
Tian Y , Wang G , Li H , et al . A novel deep learning method based on 2‑D CNNs and GRUs for permeability prediction of tight sandstone [J]. Geoenergy Science and Engineering , 2024 , 238 : 212851 .
Alao O J , Lawal M K , Dewu M B B , et al . Detection of shallow underground targets using electrical resistivity tomography and the implications in civil/environmental engineering [J]. Discover Geoscience , 2024 , 2 ( 1 ): 52 ‑ 52 .
盛丰 , 文鼎 , 熊祎玮 , 等 . 基于电阻率层析成像技术的农田土壤优先流原位动态监测 [J]. 农业工程学报 , 2021 , 37 ( 8 ): 117 ‑ 124 .
Sheng F , Wen D , Xiong Y W , et al . In‑situ monitoring of preferential soil water flow with electrical resistivity tomography technology [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE) , 2021 , 37 ( 8 ): 117 ‑ 124 . (in Chinese)
Perdomo S , Kruse E E , Ainchil J E . Estimation of hydraulic parameters using electrical resistivity tomography (ERT) and empirical laws in a semi‐confined aquifer [J]. Near Surface Geophysics , 2018 , 16 ( 6 ): 627 ‑ 641 .
González J A M , Comte J C , Legchenko A , et al . Quantification of groundwater storage heterogeneity in weathered/fractured basement rock aquifers using electrical resistivity tomography: Sensitivity and uncertainty associated with petrophysical modelling [J]. Journal of Hydrology , 2021 , 593 : 125637 .
刘征宇 , 庞永昊 , 王传武 , 等 . 基于不完整边界先验信息的跨孔电阻率CT反演方法研究 [J]. 岩土工程学报 , 2019 , 41 ( 6 ): 1124 ‑ 1132 .
Liu Z Y , Pang Y H , Wang C W , et al . Cross‑hole resistivity inversion method constrained by prior information of incomplete boundary [J]. Chinese Journal of Geotechnical Engineering , 2019 , 41 ( 6 ): 1124 ‑ 1132 . (in Chinese)
刘征宇 , 李术才 , 刘斌 , 等 . 基于距离加权约束算法的围岩三维电阻率CT反演成像研究 [J]. 岩土工程学报 , 2017 , 39 ( 4 ): 652 ‑ 661 .
Liu Z Y , Li S C , Liu B , et al . 3D cross‑hole resistivity inversion imaging of surrounding rock based on distance weighting constraint algorithm [J]. Chinese Journal of Geotechnical Engineering , 2017 , 39 ( 4 ): 652 ‑ 661 . (in Chinese)
Liu X , Shen J , Yang M , et al . Subsurface characterization of a construction site in Nanjing, China using ERT and CPTU methods [J]. Engineering Geology , 2022 , 299 : 106563 .
Cromwell E , Shuai P , Jiang P , et al . Estimating watershed subsurface permeability from stream discharge data using deep neural networks [J]. Frontiers in Earth Science , 2021 , 9 : 613011 .
Hou X , Hu R , Yeh T C J , et al . A short‐term pumping strategy for hydraulic tomography based on the successive linear estimator [J]. Water Resources Research , 2023 , 59 ( 4 ): e2022WR033831 .
慎国强 , 朱磊 , 王希萍 , 等 . 投影卷积神经网络固、液相钾盐矿层预测 [J]. 地质学报 , 2024 , 98 ( 10 ): 2946 ‑ 2955 .
Shen G Q , Zhu L , Wang X P , et al . Prediction of solid and liquid phase potash reservoirs with projected convolutional neural network [J]. Acta Geologica Sinica , 2024 , 98 ( 10 ): 2946 ‑ 2955 . (in Chinese)
丁莹莹 , 尹尚先 , 连会青 , 等 . 基于SSA‑CG‑Attention模型的多因素采煤工作面涌水量预测 [J]. 煤田地质与勘探 , 2024 , 52 ( 4 ): 111 - 119 .
Ding Y Y , Yin S X , Lian H Q , et al . Prediction of mine water inflow along mining faces using the SSA‑CG‑Attention multifactor model [J]. Coal Geology & Exploration , 2024 , 52 ( 4 ): 111 ‑ 119 . (in Chinese)
Zhang G , Wang 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 .
0
Views
0
下载量
0
CSCD
Publicity Resources
Related Articles
Related Author
Related Institution
苏公网安备32010202012147号