Predicting permeability from 3D rock images based on CNN with physical information

Tang，Pengfei1; Zhang，Dongxiao2,3,4; Li，Heng5

2022-03-01

10.1016/j.jhydrol.2022.127473

JOURNAL OF HYDROLOGY   影响因子和分区

0022-1694

1879-2707

Permeability is one of the most important properties in subsurface flow problems, which measures the ability of rocks to transmit fluid. Normally, permeability is determined through experiments and numerical simulations, both of which are time-consuming. In this paper, we propose a new effective method based on convolutional neural networks with physical information (CNN) to rapidly evaluate rock permeability from its three-dimensional (3D) image. In order to obtain sufficient reliable labeled data, rock image reconstruction is utilized to generate sufficient samples based on the Joshi-Quiblier-Adler method. Next, the corresponding permeability is calculated using the Lattice Boltzmann method. We compare the prediction performance of CNN and convolutional neural networks (CNNs). The results demonstrate that CNN achieves superior performance, especially in the case of a small dataset and an out-of-range problem. Moreover, the performance of both CNN and CNN is greatly improved combined with transfer learning in the case of an out-of-range problem. This opens novel pathways for rapidly predicting permeability in subsurface applications.

Deep learning Out-of-range problem Permeability prediction Physical information Small dataset

SCI ; EI

英语

通讯

Shenzhen Key Laboratory of Natural Gas Hydrates[ZDSYS20200421111201738]

Engineering ; Geology ; Water Resources

Engineering, Civil ; Geosciences, Multidisciplinary ; Water Resources

WOS:000752810500003

ELSEVIER

20220411495068

Convolution ; Deep learning ; Forecasting ; Image reconstruction ; Mechanical permeability ; Rocks

Ergonomics and Human Factors Engineering:461.4 ; Information Theory and Signal Processing:716.1

ENGINEERING

2-s2.0-85123027892

Scopus

1.Department of Energy and Resources Engineering,College of Engineering,Peking University,Beijing,100871,China
2.Shenzhen Key Laboratory of Natural Gas Hydrates,Southern University of Science and Technology,Shenzhen,518055,China
3.School of Environmental Science and Engineering,Southern University of Science and Technology,Shenzhen,518055,China
4.Intelligent Energy Laboratory,Peng Cheng Laboratory,Shenzhen,518000,China
5.School of Earth Resources,China University of Geosciences,Wuhan,430074,China

Tang，Pengfei,Zhang，Dongxiao,Li，Heng. Predicting permeability from 3D rock images based on CNN with physical information[J]. JOURNAL OF HYDROLOGY,2022,606.

Tang，Pengfei,Zhang，Dongxiao,&Li，Heng.(2022).Predicting permeability from 3D rock images based on CNN with physical information.JOURNAL OF HYDROLOGY,606.

Tang，Pengfei,et al."Predicting permeability from 3D rock images based on CNN with physical information".JOURNAL OF HYDROLOGY 606(2022).