Reliable and Practical Inverse Modeling of Natural‐State Geothermal Systems Using Physics‐Informed Neural Networks: Three‐Dimensional Model Construction and Assimilation With Magnetotelluric Data
Journal of Geophysical Research: Machine Learning and Computation · 2025-09 · 2 (3)
DOI: 10.1029/2025jh000683 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Geophysical Research: Machine Learning and Computation, online erschienen am 23.07.2025. Verfasst von K. Ishitsuka und 10 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 129 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: New Energy and Industrial Technology Development Organization, Japan Society for the Promotion of Science.
Abstract (Originalsprache)
Inverse modeling of geothermal systems is crucial for geothermal resource development and understanding of underground thermal structures; however, conventional modeling by calibrating numerical simulation has been known to have a high computational load and pitfalls to fall in local minima. Physics‐informed neural networks (PINNs) are an emerging method for inverse numerical modeling of partial differential equations using deep learning. Yet, in particular, 3D modeling in Earth science requires efficient training in physical laws and verification of effectiveness with limited data. This study examined the effectiveness of PINNs in the inverse modeling of natural‐state geothermal systems by predicting the 3D temperature, pressure, and permeability structure of the Kakkonda geothermal field in Japan based on synthetic well data. We introduced coefficient annealing into the loss function of the neural network to efficiently train the physical laws. Our results demonstrated that temperatures and pressures modeled by the PINN outperformed those by data‐driven machine learning in terms of prediction accuracy and physical interpretability with a limited amount of well data. The effects of setting boundary conditions and optimizers were also investigated to improve the method's applicability. We demonstrated the effectiveness of the transfer learning of pretrained networks in improving the permeability field's accuracy. This study further demonstrated that training the network by assimilating magnetotelluric data and well data improved the prediction accuracy of permeability and physical reliability.
Förderung
- New Energy and Industrial Technology Development Organization
- Japan Society for the Promotion of Science
Zitieren
APA
Ishitsuka, K., Ishizu, K., Watanabe, N., Yamaya, Y., Suzuki, A., Bandai, T., Ohta, Y., Mogi, T., Asanuma, H., Kajiwara, T., & Sugimoto, T. (2025). Reliable and Practical Inverse Modeling of Natural‐State Geothermal Systems Using Physics‐Informed Neural Networks: Three‐Dimensional Model Construction and Assimilation With Magnetotelluric Data. Journal of Geophysical Research: Machine Learning and Computation, 2(3). https://doi.org/10.1029/2025jh000683
Vancouver
Ishitsuka K, Ishizu K, Watanabe N, Yamaya Y, Suzuki A, Bandai T, et al. Reliable and Practical Inverse Modeling of Natural‐State Geothermal Systems Using Physics‐Informed Neural Networks: Three‐Dimensional Model Construction and Assimilation With Magnetotelluric Data. Journal of Geophysical Research: Machine Learning and Computation. 2025;2(3). doi:10.1029/2025jh000683
BibTeX
@article{Ishitsuka2025,
author = {Ishitsuka, K. and Ishizu, K. and Watanabe, N. and Yamaya, Y. and Suzuki, A. and Bandai, T. and Ohta, Y. and Mogi, T. and Asanuma, H. and Kajiwara, T. and Sugimoto, T.},
title = {Reliable and Practical Inverse Modeling of Natural‐State Geothermal Systems Using Physics‐Informed Neural Networks: Three‐Dimensional Model Construction and Assimilation With Magnetotelluric Data},
journal = {Journal of Geophysical Research: Machine Learning and Computation},
year = {2025},
volume = {2},
number = {3},
doi = {10.1029/2025jh000683}
}
EUVIMED