Abstract:The wave equation is an important physical partial differential equation, and in recent years, deep learning has shown promise in accelerating or replacing traditional numerical methods for solving it. However, existing deep learning methods suffer from high data acquisition costs, low training efficiency, and insufficient generalization capability for boundary conditions. To address these issues, this paper proposes an unsupervised learning method for the wave equation based on finite difference residual constraints. We construct a novel finite difference residual constraint based on structured grids and finite difference methods, as well as an unsupervised training strategy, enabling convolutional neural networks to train without data and predict the forward propagation process of waves. Experimental results show that finite difference residual constraints have advantages over physics-informed neural networks (PINNs) type physical information constraints, such as easier fitting, lower computational costs, and stronger source term generalization capability, making our method more efficient in training and potent in application.

What problem does this paper attempt to address?

The problems that this paper attempts to solve are several key issues existing in the existing deep - learning methods when solving the wave equation: 1. **High cost of data set acquisition**: Traditional data - driven deep - learning methods require a large amount of training data, and this data is usually difficult to obtain and costly. 2. **Low training efficiency**: The existing deep - learning methods have problems of high computational complexity and long training time during the training process. 3. **Insufficient generalization ability of boundary conditions**: When the boundary conditions change, the model needs to be retrained, resulting in a limited application range. To solve these problems, the author proposes an unsupervised learning method for the wave equation based on Finite Difference Residual Constraints (FDRC). Specifically: - **Finite Difference Residual Constraints**: By combining the traditional finite - difference method with physics - informed learning, a novel finite - difference residual constraint is constructed, enabling the convolutional neural network to be trained without data and predict the forward process of waves. - **Unsupervised training strategy**: An unsupervised training strategy based on the training pool is proposed, which improves the training efficiency of the model and solves the problem of difficult data acquisition at the same time. - **Improved generalization ability of boundary conditions**: Compared with Physics - Informed Neural Networks (PINNs), the FDRC method has better adaptability to changes in boundary conditions and does not need to retrain the model. The experimental results show that this method is superior to the existing PINNs - like methods in terms of fitting speed, computational cost, and source - term generalization ability, and has higher training efficiency and application potential.

Unsupervised Learning Method for the Wave Equation Based on Finite Difference Residual Constraints Loss

An Unsupervised Deep Learning Approach for the Wave Equation Inverse Problem

Non-equispaced Fourier Neural Solvers for PDEs

Solving Partial Differential Equations Using Deep Learning and Physical Constraints

Data Generation-based Operator Learning for Solving Partial Differential Equations on Unbounded Domains

Explorations of certain nonlinear waves of the Boussinesq and Camassa–Holm equations using physics-informed neural networks

Physics-driven Learning of the Steady Navier-Stokes Equations using Deep Convolutional Neural Networks

An Improved PINN Algorithm for Shallow Water Equations Driven by Deep Learning

Physics informed deep learning for computational elastodynamics without labeled data

A robust numerical method based on deep learning operator for solving the 2D acoustic wave equation

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

Numerical solution for high-dimensional partial differential equations based on deep learning with residual learning and data-driven learning

A Robust Numerical Method Based on a Deep-Learning Operator for Solving the 2D Acoustic Wave Equation

Physical informed neural networks with soft and hard boundary constraints for solving advection-diffusion equations using Fourier expansions

Soliton, Breather and Rogue Wave Solutions for Solving the Nonlinear Schrödinger Equation Using a Deep Learning Method with Physical Constraints

Wavefield solutions from machine learned functions

A deep learning operator-based numerical scheme method for solving 1-D wave equations

Study on a Poisson's Equation Solver Based on Deep Learning Technique.

A Stabilized Physics Informed Neural Networks Method for Wave Equations

A fully differentiable GNN-based PDE Solver: With Applications to Poisson and Navier-Stokes Equations