Deep learning of dynamic subsurface flow via theory-guided generative adversarial network

He，Tianhao1; Zhang，Dongxiao2,3

2021-10-01

10.1016/j.jhydrol.2021.126626

JOURNAL OF HYDROLOGY   影响因子和分区

0022-1694

Generative adversarial network (GAN) has been shown to be useful in various applications, such as image recognition, text processing and scientific computing, due its strong ability to learn complex data distributions. However, the ability of standard GAN to process dynamic data is limited. In this study, a theory-guided generative adversarial network (TgGAN) is proposed to solve dynamic partial differential equations (PDEs). Different from standard GANs, the training term is no longer the true data and the generated data, but rather their residuals. In addition, such theories as governing equations and other physical constraints are encoded into the loss function of the generator to ensure that the prediction does not only honor the training data, but also obey these theories. TgGAN is proposed for dynamic subsurface flow with heterogeneous model parameters, and the data at each time step are treated as a two-dimensional image. In this study, several numerical cases are introduced to test the performance of the TgGAN. Predicting the future response, label-free learning and learning from noisy data can be realized easily by the TgGAN model, and the effects of the number of training data and the collocation points are also discussed. In order to improve the efficiency of TgGAN, the transfer learning algorithm is also employed. Moreover, the sensitivity of TgGAN to the hydraulic conductivity field is studied. Numerical results demonstrate that the TgGAN model is both robust and reliable for deep learning of dynamic PDEs.

Deep learning Dynamic subsurface flow Generative adversarial network Theory-guided

SCI ; EI

英语

通讯

WOS:000695816300043

20212810622475

Character recognition ; Computation theory ; Computerized tomography ; Deep learning ; Learning algorithms ; Transfer learning

Ergonomics and Human Factors Engineering:461.4 ; Computer Theory, Includes Formal Logic, Automata Theory, Switching Theory, Programming Theory:721.1 ; Artificial Intelligence:723.4 ; Machine Learning:723.4.2 ; Computer Applications:723.5

ENGINEERING

2-s2.0-85109450552

Scopus

1.College of Engineering,Peking University,Beijing,100871,China
2.School of Environmental Science and Engineering,Southern University of Science and Technology,Shenzhen,518055,China
3.Intelligent Energy Laboratory,Peng Cheng Laboratory,Shenzhen,518000,China

He，Tianhao,Zhang，Dongxiao. Deep learning of dynamic subsurface flow via theory-guided generative adversarial network[J]. JOURNAL OF HYDROLOGY,2021,601.

He，Tianhao,&Zhang，Dongxiao.(2021).Deep learning of dynamic subsurface flow via theory-guided generative adversarial network.JOURNAL OF HYDROLOGY,601.

He，Tianhao,et al."Deep learning of dynamic subsurface flow via theory-guided generative adversarial network".JOURNAL OF HYDROLOGY 601(2021).