A critical review of physics-informed machine learning applications in subsurface energy systems

A Latrach, ML Malki, M Morales, M Mehana… - Geoenergy Science and …, 2024 - Elsevier
Abstract Machine learning has emerged as a powerful tool in various fields, including
computer vision, natural language processing, and speech recognition. It can unravel …

U-no: U-shaped neural operators

MA Rahman, ZE Ross, K Azizzadenesheli - arxiv preprint arxiv …, 2022 - arxiv.org
Neural operators generalize classical neural networks to maps between infinite-dimensional
spaces, eg, function spaces. Prior works on neural operators proposed a series of novel …

Review of physics-informed machine-learning inversion of geophysical data

GT Schuster, Y Chen, S Feng - Geophysics, 2024 - library.seg.org
We review five types of physics-informed machine-learning (PIML) algorithms for inversion
and modeling of geophysical data. Such algorithms use the combination of a data-driven …

3D elastic wave propagation with a factorized Fourier neural operator (F-FNO)

F Lehmann, F Gatti, M Bertin, D Clouteau - Computer Methods in Applied …, 2024 - Elsevier
Numerical simulations are computationally demanding in three-dimensional (3D) settings
but they are often required to accurately represent physical phenomena. Neural operators …

Deep neural Helmholtz operators for 3D elastic wave propagation and inversion

C Zou, K Azizzadenesheli, ZE Ross… - arxiv preprint arxiv …, 2023 - arxiv.org
Numerical simulations of seismic wave propagation in heterogeneous 3D media are central
to investigating subsurface structures and understanding earthquake processes, yet are …

Lemon: Learning to learn multi-operator networks

J Sun, Z Zhang, H Schaeffer - arxiv preprint arxiv:2408.16168, 2024 - arxiv.org
Single-operator learning involves training a deep neural network to learn a specific operator,
whereas recent work in multi-operator learning uses an operator embedding structure to …

Out-of-distributional risk bounds for neural operators with applications to the Helmholtz equation

JAL Benitez, T Furuya, F Faucher, A Kratsios… - Journal of …, 2024 - Elsevier
Despite their remarkable success in approximating a wide range of operators defined by
PDEs, existing neural operators (NOs) do not necessarily perform well for all physics …

Predictions of transient vector solution fields with sequential deep operator network

J He, S Kushwaha, J Park, S Koric, D Abueidda… - Acta Mechanica, 2024 - Springer
The deep operator network (DeepONet) structure has shown great potential in
approximating complex solution operators with low generalization errors. Recently, a …

HOSSNet: An efficient physics-guided neural network for simulating micro-crack propagation

S Chen, S Feng, Y Huang, Z Lei, X Jia, Y Lin… - Computational Materials …, 2024 - Elsevier
Abstract The Hybrid Optimization Software Suite (HOSS), which combines the finite-discrete
element method (FDEM), is an advanced approach for simulating high-fidelity fracture and …

Transfer learning Fourier neural operator for solving parametric frequency-domain wave equations

Y Wang, H Zhang, C Lai, X Hu - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Fourier neural operator (FNO) is a recently proposed data-driven scheme to approximate the
implicit operators characterized by partial differential equations (PDEs) between functional …