4.6 Article

A versatile framework to solve the Helmholtz equation using physics-informed neural networks

Related references

Note: Only part of the references are listed.
Article Geochemistry & Geophysics

Frequency domain full-waveform inversion in a fluid-saturated poroelastic medium

Qingjie Yang et al.

Summary: The study introduces the method of inversion of subsurface properties using poroelastic theory, and predicts trade-offs among parameters based on sensitivity kernels and radiation patterns. Numerical experiments verify the feasibility and effectiveness of this method, while various factors influencing inversion results are discussed.

GEOPHYSICAL JOURNAL INTERNATIONAL (2021)

Article Engineering, Multidisciplinary

On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networks

Sifan Wang et al.

Summary: PINNs show promise in integrating physical models with observational data, but struggle with high-frequency or multi-scale features. Through NTK theory, this work investigates the limitations of PINNs and proposes novel architectures to address these challenges.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2021)

Article Computer Science, Interdisciplinary Applications

PINNeik: Eikonal solution using physics-informed neural networks

Umair bin Waheed et al.

Summary: The proposed algorithm based on physics-informed neural networks offers a high traveltime accuracy for a wide range of applications in seismology. It leverages machine learning techniques like transfer learning and surrogate modeling to speed up computations for updated velocity models and source locations. The method's flexibility in incorporating various physical properties and improving convergence rate makes it an efficient forward modeling engine for seismological applications.

COMPUTERS & GEOSCIENCES (2021)

Article Geochemistry & Geophysics

EikoNet: Solving the Eikonal Equation With Deep Neural Networks

Jonathan D. Smith et al.

Summary: EikoNet is a deep learning approach proposed to solve the Eikonal equation, allowing for rapid determination of travel time between any two points in a continuous 3-D domain. By exploiting the differentiability of neural networks, the method calculates spatial gradients analytically without needing solutions from a finite-difference algorithm, and has been rigorously tested on velocity models and sampling methods.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2021)

Article Geochemistry & Geophysics

Solving the frequency-domain acoustic VTI wave equation using physics-informed neural networks

Chao Song et al.

Summary: Frequency-domain wavefield solutions for the anisotropic acoustic wave equation can help describe the anisotropic nature of the Earth, but inverting the impedance matrix for such solutions can be computationally expensive. To address this challenge, the study proposes using physics-informed neural networks (PINNs) to obtain wavefield solutions for an acoustic wave equation in transversely isotropic (TI) media with a vertical axis of symmetry (VTI). By utilizing automatic differentiation in PINNs, numerical dispersion artefacts can be avoided, providing more accurate solutions.

GEOPHYSICAL JOURNAL INTERNATIONAL (2021)

Article Engineering, Aerospace

Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows

Nils Thuerey et al.

AIAA JOURNAL (2020)

Article Geochemistry & Geophysics

Seismic ground-roll noise attenuation using deep learning

Harpreet Kaur et al.

GEOPHYSICAL PROSPECTING (2020)

Article Geosciences, Multidisciplinary

Physics informed machine learning: Seismic wave equation

Sadegh Karimpouli et al.

GEOSCIENCE FRONTIERS (2020)

Article Materials Science, Characterization & Testing

Physics-Informed Neural Network for Ultrasound Nondestructive Quantification of Surface Breaking Cracks

Khemraj Shukla et al.

JOURNAL OF NONDESTRUCTIVE EVALUATION (2020)

Article Geochemistry & Geophysics

Deep learning for fast simulation of seismic waves in complex media

Ben Moseley et al.

SOLID EARTH (2020)

Article Computer Science, Interdisciplinary Applications

Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

M. Raissi et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2019)

Article Mechanics

Deep learning of vortex-induced vibrations

Maziar Raissi et al.

JOURNAL OF FLUID MECHANICS (2019)

Article Geochemistry & Geophysics

An efficient wave extrapolation method for anisotropic media with tilt

Umair bin Waheed et al.

GEOPHYSICAL PROSPECTING (2015)

Article Geochemistry & Geophysics

Data space reduction, quality assessment and searching of seismograms: autoencoder networks for waveform data

Andrew P. Valentine et al.

GEOPHYSICAL JOURNAL INTERNATIONAL (2012)

Article Geochemistry & Geophysics

Stable P-wave modeling for reverse-time migration in tilted TI media

Eric Duveneck et al.

GEOPHYSICS (2011)

Article Computer Science, Artificial Intelligence

A Survey on Transfer Learning

Sinno Jialin Pan et al.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2010)

Article Geochemistry & Geophysics

Automatic picking of P and S phases using a neural tree

S. Gentili et al.

JOURNAL OF SEISMOLOGY (2006)

Article Geochemistry & Geophysics

Neural networks in geophysical applications

M van der Baan et al.

GEOPHYSICS (2000)

Article Geochemistry & Geophysics

An acoustic wave equation for anisotropic media

T Alkhalifah

GEOPHYSICS (2000)