Abstract:Abstract Physics-informed neural networks (PINNs) have been used by researchers to solve partial differential equation (PDE)-constrained problems. We evaluate PINNs to solve for frequency-domain acoustic wavefields. PINNs can solely use PDEs to define the loss function for optimization without the need for labels. Partial derivatives of PDEs are calculated by mesh-free automatic differentiations. Thus, PINNs are free of numerical dispersion artifacts. It has been applied to the scattered acoustic wave equation, which relied on boundary conditions (BCs) provided by the background analytical wavefield. For a more direct implementation, we solve the nonscattered acoustic wave equation, avoiding limitations related to relying on the background homogeneous medium for BCs. Experiments support our following insights. Although solving time-domain wave equations using PINNs does not require absorbing boundary conditions (ABCs), ABCs are required to ensure a unique solution for PINNs that solve frequency-domain wave equations, because the single-frequency wavefield is not localized and contains wavefield information over the full domain. However, it is not trivial to include the ABC in the PINN implementation, so we develop an adaptive amplitude-scaled and phase-shifted sine activation function, which performs better than the previous implementations. Because there are only two outputs for the fully connected neural network (FCNN), we validate a linearly shrinking FCNN that can achieve a comparable and even better accuracy with a cheaper computational cost. However, there is a spectral bias problem, that is, PINNs learn low-frequency wavefields far more easily than higher frequencies, and the accuracy of higher frequency wavefields is often poor. Because the shapes of multifrequency wavefields are similar, we initialize the FCNN for higher frequency wavefields by that of the lower frequencies, partly mitigating the spectral bias problem. We further incorporate multiscale positional encoding to alleviate the spectral bias problem. We share our codes, data, and results via a public repository.

Mathematics-Informed Neural Network for Acoustic Scattering by Parallel Plates

Acoustic Field Visualization and Source Localization Via Physics-Informed Learning of Sparse Data with Adaptive Sampling

Multiple scattering simulation via physics-informed neural networks

Practical Aspects of Physics-Informed Neural Networks Applied to Solve Frequency-Domain Acoustic Wave Forward Problem

Wavefield solutions from machine learned functions

Deep ReLU Neural Network Emulation in High-Frequency Acoustic Scattering

Helmholtz-equation Solution in Nonsmooth Media by a Physics-Informed Neural Network Incorporating Quadratic Terms and a Perfectly Matching Layer Condition

Modeling multisource multifrequency acoustic wavefields by a multiscale Fourier feature physics-informed neural network with adaptive activation functions

A Neural Network Warm-Start Approach for the Inverse Acoustic Obstacle Scattering Problem

Boundary integrated neural networks and code for acoustic radiation and scattering

Physics and geometry informed neural operator network with application to acoustic scattering

Deep neural networks for waves assisted by the Wiener-Hopf method

Taper-based scattering formulation of the Helmholtz equation to improve the training process of Physics-Informed Neural Networks

A More General Electromagnetic Inverse Scattering Method Based on Physics-Informed Neural Network

Physics Embedded Deep Neural Network for Solving Full-Wave Inverse Scattering Problems

Solving Inverse Wave Scattering with Deep Learning

Multi-frequency wavefield modeling of acoustic VTI wave equation using physics informed neural networks

Spatial acoustic properties recovery with deep learning

Point Neuron Learning: A New Physics-Informed Neural Network Architecture

Feasibility study on solving the Helmholtz equation in 3D with PINNs

Wave simulation in non-smooth media by PINN with quadratic neural network and PML condition